(273 days)
This software is a computer-assisted reading tool designed to aid endoscopists in detecting colonic mucosal lesions (such as polyps and adenomas) in real time during standard endoscopy examinations of patients undergoing screening and surveillance endoscopic mucosal evaluations. This software is used with standard White Light Imaging (WLI) and Linked Color Imaging (LCI) endoscopy imaging. This software is not intended to replace clinical decision making.
The subject device represents application of AI technology to endoscopic images to assist in detecting the presence of potential lesions. This development greatly contributes to improving the quality of colonoscopy. In recent years, computer-aided diagnosis (CAD) systems employing AI technologies have been approved and marketed as radiological medical devices for use with computed tomography (CT), X-ray, magnetic resonance imaging (MRI), and mammogram diagnostic images. In endoscopy as well, many images for diagnosis are taken. Since increasing the polyp detection rate is also in demand, CAD systems for endoscopy are being actively developed. Against this background, the company has developed this software (EW10-EC02), a new AI-based CAD system, to support Health Care Provider (HCP) detection of large intestine polyps in colonoscopic images. EW10-EC02 detects suspected large intestine polyps in the endoscope video image in real-time.
Here's a breakdown of the acceptance criteria and study findings for the EW10-EC02 Endoscopy Support Program, based on the provided document:
1. Table of Acceptance Criteria and Reported Device Performance
The document describes two main types of studies: standalone performance testing (evaluating the algorithm only) and clinical testing (evaluating human-in-the-loop performance). The acceptance criteria for the standalone performance are explicitly stated and met, while the clinical study endpoints serve as the criteria for evaluating the device's clinical benefit when assisting human readers.
Standalone Performance Acceptance Criteria & Results:
| Item | Acceptance Criteria (Implicit, based on "achieved all criteria") | Reported Performance WLI Mode | Reported Performance LCI Mode |
|---|---|---|---|
| Sensitivity per lesion (Lesion-based sensitivity) | Exceeds a defined lower limit of the 95% CI (Specific value not provided but stated as met) | 95.1% (91.1 - 98.3% CI) | 95.5% (91.5 - 98.7% CI) |
| FP Objects/Patient (Number of FPc per Case) | (Specific criteria not numerically stated, but described as "achieved all criteria") | 1.42 (1.09 - 1.81 CI) | 0.76 (0.42 - 1.21 CI) |
| Detection Persistence (Figure 1) | (Implicit: Robust correlation of detection persistence with sensitivity and FP objects/patient) | Demonstrated strong correlation | Demonstrated strong correlation |
| Frame-level performance | (Implicit: Acceptable values for TP, TN, FP, FN, sensitivity/frame, FPR/frame) | (See Table 7 for detailed values) | (See Table 7 for detailed values) |
| ROC AUC | (Implicit: High accuracy) | 0.79 (0.77-0.80 CI) | 0.87 (0.86-0.88 CI) |
| FROC Analysis | (Implicit: Supports performance) | (See Figure 4) | (See Figure 4) |
Clinical Study Endpoints & Results (serving as criteria for human-in-the-loop):
| Endpoint | Acceptance Criteria (Implicit: Superiority for APC or meeting margins for PPV; non-inferiority for FPR) | Reported Performance (CAC group vs. CC group) | P-Value / CI |
|---|---|---|---|
| Primary Endpoints | |||
| Adenoma per colonoscopy (APC) | Superiority (CAC vs. CC) | CAC: 0.990 ± 1.610; CC: 0.849 ± 1.484 | 0.018 (Superiority met) |
| Positive predictive value (PPV) | Meeting margin of -9.56% | CAC: 48.6%; CC: 54.0% | -9.56%, -1.48% (Margin met) |
| Positive percent agreement (PPA) | (Implicit: Acceptable performance) | CAC: 60.7%; CC: 66.2% | -10.50%, -2.30% |
| Secondary Endpoints of Note | |||
| Polyp per colonoscopy (PPC) | (Implicit: Acceptable performance, P-value <0.001 noted) | CAC: 1.680 ± 2.070; CC: 1.328 ± 1.791 | <0.001 (Statistically significant) |
| Serrated Lesions per Colonoscopy (SLPC) | (Implicit: Acceptable performance, P-value <0.001 noted) | CAC: 0.171 ± 0.502; CC: 0.130 ± 0.478 | 0.094 |
| Serrated Lesions Detection Rate (SLDR) | (Implicit: Acceptable performance, P-value 0.027 noted) | CAC: 13.0%; CC: 10.2% | 0.157 |
| False Positive Rate (FPR) | Non-inferiority | CAC: 17.6%; CC: 15.0% | 1.39% - 7.82% (Non-inferiority demonstrated) |
| True Histology Rate (THR) | (Implicit: Acceptable performance) | CAC: 57.0%; CC: 62.3% | -10.3%, -2.06% |
2. Sample Size Used for the Test Set and Data Provenance
Standalone Performance Testing (Algorithm only):
- Sample Size: 149 colonoscopy videos for WLI mode and 144 colonoscopy videos for LCI mode.
- WLI: 119 patients with lesions, 30 patients without lesions.
- LCI: 114 patients with lesions, 30 patients without lesions.
- Data Provenance:
- Country of origin: Implied to be primarily of Asian descent (100% Asian specified in patient demographics, Table 2). While not explicitly stated, this suggests data from an Asian country (likely Japan, given the manufacturer's location).
- Retrospective or Prospective: Retrospective, as these were "colonoscopy videos" used for evaluation, not real-time clinical use for testing.
Clinical Testing (Human-in-the-loop):
- Sample Size: 1,031 subjects analyzed (out of 1,166 enrolled).
- CAC (Computer Assisted Colonoscopy) group: 509 subjects
- CC (Conventional Colonoscopy) group: 522 subjects
- Data Provenance:
- Country of origin: United States (conducted at 12 centers in the United States).
- Retrospective or Prospective: Prospective, randomized controlled trial.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications
Standalone Performance Testing:
- The document states that "lesion detection was performed and evaluated at the lesion level (object level) in the actual clinical cases". However, it does not specify the number of experts used to establish the ground truth for the standalone test set. It also does not detail their qualifications (e.g., years of experience, board certification). The data originated from "actual clinical cases," implying that the ground truth was derived from the in-vivo findings and subsequent histopathology from those cases.
Clinical Testing:
- The ground truth for this study was established by histopathology of resected polyps. This is considered the gold standard for lesion characterization. Endoscopists performed the resections during the colonoscopies. The document does not specify the number of pathologists or their qualifications who reviewed the specimens, but typically, this is done by qualified histopathologists.
4. Adjudication Method for the Test Set
Standalone Performance Testing:
- The document does not explicitly describe an adjudication method for the ground truth of the standalone test set. It states that sensitivity per lesion was the primary endpoint and refers to evaluation values calculated per frame and per case, based on detection persistence. This suggests a direct comparison to the established ground truth without a specific expert adjudication process for discrepancies.
Clinical Testing:
- The primary ground truth for the clinical study was histopathology. For endpoints like APC, PPV, and PPA, histopathology results directly determined the presence and type of lesions. There is no mention of an explicitly described adjudication method (like 2+1 or 3+1 expert review) for the clinical study results, as the histopathological diagnosis serves as the definitive ground truth for the resected tissue.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done, and the Effect Size of Improvement with AI vs. without AI Assistance
- A MRMC study was done for the clinical performance evaluation. This was a multi-center, prospective, randomized controlled trial comparing Computer Assisted Colonoscopy (CAC group) with AI assistance to Conventional Colonoscopy (CC group) without AI assistance.
- Effect Size of Improvement (How human readers improve with AI vs. without AI assistance):
- Adenoma Per Colonoscopy (APC): The mean number of adenomas per colonoscopy increased from 0.849 in the Conventional Colonoscopy (CC) group to 0.990 in the Computer Assisted Colonoscopy (CAC) group. This represents an absolute difference of 0.141 (95% CI: 0.01, 0.28), which was statistically significant (p=0.018). This demonstrates that AI assistance led to an increase in the detection of adenomas per patient.
- Polyp Per Colonoscopy (PPC): The mean number of polyps per colonoscopy increased from 1.328 in the CC group to 1.680 in the CAC group. This was also statistically significant (p<0.001).
- Serrated Lesions Per Colonoscopy (SLPC): While not statistically significant at 0.05, the mean SLPC increased from 0.130 in CC to 0.171 in CAC (p=0.094).
- Serrated Lesions Detection Rate (SLDR): The rate increased from 29.1% in CC to 35.6% in CAC, with a p-value of 0.027, indicating a statistically significant improvement in the detection rate of serrated lesions.
6. If a Standalone (Algorithm Only without Human-in-the-Loop Performance) was done
- Yes, a standalone performance testing was done. This is detailed under the "Non-clinical Performance Testing" section. The evaluation assessed the device's object-level, frame-level, and overall algorithmic performance.
7. The Type of Ground Truth Used
Standalone Performance Testing:
- The ground truth was established by annotations of lesions identified in the colonoscopy videos, with the ultimate confirmation likely tied to the patient's pathology results for those lesions. It refers to "actual clinical cases" where "lesion detection was performed and evaluated at the lesion level (object level)".
Clinical Testing:
- The definitive ground truth for the clinical study was histopathology (pathology results) of resected polyps. This is explicitly stated for the primary endpoints (APC, PPV, PPA and many secondary endpoints like ADR, SLPC, etc.) which were based on "histologically confirmed" findings.
8. The Sample Size for the Training Set
- The document does not explicitly state the sample size used for the training set of the AI algorithm. It only discusses the dataset used for standalone performance testing (test set) and the clinical study population.
9. How the Ground Truth for the Training Set was Established
- The document does not explicitly describe how the ground truth for the training set was established. It mentions the EW10-EC02 utilizes an "artificial intelligence-based algorithm to perform the polyp detection function." Typically, for such AI systems, the training data is extensively annotated by medical experts (e.g., endoscopists, pathologists) to provide the ground truth for the algorithm to learn from. However, the specific process (e.g., number of annotators, their qualifications, adjudication methods for training data) is not detailed in this submission.
{0}------------------------------------------------
Image /page/0/Picture/0 description: The image shows the logo of the U.S. Food and Drug Administration (FDA). On the left is the Department of Health & Human Services logo. To the right of that is the FDA logo in blue, with the words "U.S. FOOD & DRUG ADMINISTRATION" in blue as well. The FDA is a federal agency responsible for regulating and supervising the safety of food, drugs, and other products.
December 15, 2023
FUJIFILM Corporation % Kotei Aoki Manager, Regulatory Affairs FUJIFILM Healthcare Americas Corporation 81 Hartwell Avenue, Suite 300 Lexington, MA 02421
Re: K230751
Trade/Device Name: EW10-EC02 Endoscopy Support Program Regulation Number: 21 CFR 876.1520 Regulation Name: Gastrointestinal Lesion Software Detection System Regulatory Class: Class II Product Code: ONP Dated: November 16, 2023 Received: November 16, 2023
Dear Kotei Aoki:
We have reviewed your section 510(k) premarket notification of intent to market the device referenced above and have determined the device is substantially equivalent (for the indications for use stated in the enclosure) to legally marketed predicate devices marketed in interstate commerce prior to May 28, 1976, the enactment date of the Medical Device Amendments, or to devices that have been reclassified in accordance with the provisions of the Federal Food, Drug, and Cosmetic Act (the Act) that do not require approval of a premarket approval application (PMA). You may, therefore, market the device, subject to the general controls provisions of the Act. Although this letter refers to your product as a device, please be aware that some cleared products may instead be combination products. The 510(k) Premarket Notification Database available at https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn.cfm identifies combination product submissions. The general controls provisions of the Act include requirements for annual registration, listing of devices, good manufacturing practice, labeling, and prohibitions against misbranding and adulteration. Please note: CDRH does not evaluate information related to contract liability warranties. We remind you, however, that device labeling must be truthful and not misleading.
If your device is classified (see above) into either class II (Special Controls) or class III (PMA), it may be subject to additional controls. Existing major regulations affecting your device can be found in the Code of Federal Regulations, Title 21, Parts 800 to 898. In addition, FDA may publish further announcements concerning your device in the Federal Register.
Additional information about changes that may require a new premarket notification are provided in the FDA guidance documents entitled "Deciding When to Submit a 510(k) for a Change to an Existing Device"
{1}------------------------------------------------
(https://www.fda.gov/media/99812/download) and "Deciding When to Submit a 510(k) for a Software Change to an Existing Device" (https://www.fda.gov/media/99785/download).
Your device is also subject to, among other requirements, the Quality System (QS) regulation (21 CFR Part 820), which includes, but is not limited to, 21 CFR 820.30. Design controls; 21 CFR 820.90. Nonconforming product; and 21 CFR 820.100, Corrective and preventive action. Please note that regardless of whether a change requires premarket review. the OS regulation requires device manufacturers to review and approve changes to device design and production (21 CFR 820.30 and 21 CFR 820.70) and document changes and approvals in the device master record (21 CFR 820.181).
Please be advised that FDA's issuance of a substantial equivalence determination does not mean that FDA has made a determination that your device complies with other requirements of the Act or any Federal statutes and regulations administered by other Federal agencies. You must comply with all the Act's requirements, including, but not limited to: registration and listing (21 CFR Part 807); labeling (21 CFR Part 801); medical device reporting of medical device-related adverse events) (21 CFR Part 803) for devices or postmarketing safety reporting (21 CFR Part 4, Subpart B) for combination products (see https://www.fda.gov/combination-products/guidance-regulatory-information/postmarketing-safety-reportingcombination-products); good manufacturing practice requirements as set forth in the quality systems (QS) regulation (21 CFR Part 820) for devices or current good manufacturing practices (21 CFR Part 4, Subpart A) for combination products; and, if applicable, the electronic product radiation control provisions (Sections 531-542 of the Act); 21 CFR Parts 1000-1050.
Also, please note the regulation entitled, "Misbranding by reference to premarket notification" (21 CFR 807.97). For questions regarding the reporting of adverse events under the MDR regulation (21 CFR Part 803), please go to https://www.fda.gov/medical-device-safety/medical-device-reportingmdr-how-report-medical-device-problems.
For comprehensive regulatory information about mediation-emitting products, including information about labeling regulations, please see Device Advice (https://www.fda.gov/medicaldevices/device-advice-comprehensive-regulatory-assistance) and CDRH Learn (https://www.fda.gov/training-and-continuing-education/cdrh-learn). Additionally, you may contact the Division of Industry and Consumer Education (DICE) to ask a question about a specific regulatory topic. See the DICE website (https://www.fda.gov/medical-device-advice-comprehensive-regulatoryassistance/contact-us-division-industry-and-consumer-education-dice) for more information or contact DICE by email (DICE@fda.hhs.gov) or phone (1-800-638-2041 or 301-796-7100).
{2}------------------------------------------------
Sincerely,
Shanil P. Haugen -S
Shanil P. Haugen, Ph.D. Assistant Director DHT3A: Division of Renal, Gastrointestinal, Obesity and Transplant Devices OHT3: Office of Gastrorenal, ObGyn, General Hospital, and Urology Devices Office of Product Evaluation and Quality Center for Devices and Radiological Health
Enclosure
{3}------------------------------------------------
Indications for Use
510(k) Number (if known) K230751
Device Name Endoscopy Support Program EW10-EC02
Indications for Use (Describe)
This software is a computer-assisted reading tool designed to aid endoscopists in detecting colonic mucosal lesions (such as polyps and adenomas) in real time during standard endoscopy examinations of patients undergoing screening and surveillance endoscopic mucosal evaluations. This software is used with standard White Light Imaging (WLI) and Linked Color Imaging (LCI) endoscopy imaging. This software is not intended to replace clinical decision making.
| Type of Use (Select one or both, as applicable) |
|---|
| ------------------------------------------------- |
| Prescription Use (Part 21 CFR 801 Subpart D) | ☑ |
|---|---|
| Over-The-Counter Use (21 CFR 801 Subpart C) | ☐ |
CONTINUE ON A SEPARATE PAGE IF NEEDED.
This section applies only to requirements of the Paperwork Reduction Act of 1995.
DO NOT SEND YOUR COMPLETED FORM TO THE PRA STAFF EMAIL ADDRESS BELOW.
The burden time for this collection of information is estimated to average 79 hours per response, including the time to review instructions, search existing data sources, gather and maintain the data needed and complete and review the collection of information. Send comments regarding this burden estimate or any other aspect of this information collection, including suggestions for reducing this burden, to:
Department of Health and Human Services Food and Drug Administration Office of Chief Information Officer Paperwork Reduction Act (PRA) Staff PRAStaff(@fda.hhs.gov
"An agency may not conduct or sponsor, and a person is not required to respond to, a collection of information unless it displays a currently valid OMB number."
{4}------------------------------------------------
ട. 510(k) Summary
- a. Company Name, Address: FUJFILM Corporation 798 Miyanodai Kaisei-Machi Ashigarakami-Gun, Kanagawa, 258-8538, Japan
- b. Contact: Kotei Aoki Manager, Regulatory Affairs E-Mail: kotei.aoki@fujifilm.com Telephone: (765) 246-2931
- c. Date prepared December 15, 2023
| d. Subject Device | |
|---|---|
| 510(k) Applicant/Owner: | FUJIFILM Corporation |
| Device Name: | EW10-EC02 Endoscopy Support Program |
| Common Name: | EW10-EC02 |
| Classification Product Code: | QNP |
| Device Class: | Class II |
| Regulation Number: | 21 CFR 876.1520 |
| Regulation Description: | Gastrointestinal Lesion Software Detection System |
| Review Panel: | Gastroenterology/Urology |
{5}------------------------------------------------
e. Predicate Device
The EW10-EC02 Endoscopy Support Program is substantially equivalent to:
| 510(k) Number: | K211951 |
|---|---|
| Applicant: | Cosmo Artificial Intelligence - AI Ltd |
| Device Name: | GI Genius |
| Common Name: | GI Genius |
| Classification Product Code: | QNP |
| Device Class: | Class II |
| Regulation Number: | 21 CFR 876.1520 |
| Regulation Description: | Gastrointestinal Lesion Software DetectionSystem |
| Review Panel: | Gastroenterology/Urology |
Device Description f.
There is an increasing interest in the application of artificial intelligence (AI) in health care to improve disease diagnosis, management, and the development of effective therapies. The remarkable development of AI from deep learning in recent years has increased the possibility of machines assisting assessments by human visual observation. The subject device represents application of AI technology to endoscopic images to assist in detecting the presence of potential lesions. This development greatly contributes to improving the quality of colonoscopy.
In recent years, computer-aided diagnosis (CAD) systems employing AI technologies have been approved and marketed as radiological medical devices for use with computed tomography (CT), X-ray, magnetic resonance imaging (MRI), and mammogram diagnostic images. In endoscopy as well, many images for diagnosis are taken. Since increasing the polyp detection rate is also in demand, CAD systems for endoscopy are being actively developed.
Against this background, the company has developed this software (EW10-EC02), a new AI-based CAD system, to support Health Care Provider (HCP) detection of large intestine polyps in colonoscopic images. EW10-EC02 detects suspected large intestine polyps in the endoscope video image in real-time.
g. Intended Use / Indications for Use
This software is a computer-assisted reading tool designed to aid endoscopists in detecting colonic mucosal lesions (such as polyps and adenomas) in real time during
{6}------------------------------------------------
standard endoscopy examinations of patients undergoing screening and surveillance endoscopic mucosal evaluations. This software is used with standard WLI (White Light Imaging) and LCI (Linked Color Imaging) endoscopy imaging. This software is not intended to replace clinical decision making.
h. Statement of Substantial Equivalence
A comparison of the technological characteristics between the subject and predicate device is provided in Table 1 below. EW10-EC02 and GI Genius (K211951) share the same intended use and similar indications and technological characteristics. The differences in indications and technological characteristics between the subject and predicate devices do not raise new concerns regarding safety and effectiveness as demonstrated by the non-clinical and clinical performance evaluation results. Therefore, the Company believes that GI Genius is an appropriate predicate device for EW10-EC02, and EW10-EC02 can be submitted as a 510(k) under product code QNP.
Comparison Table i.
| Table 1: Comparison of EW10-EC02 to GI Genius | |||
|---|---|---|---|
| -- | -- | ----------------------------------------------- | -- |
| Feature | EW10-EC02Endoscopy Support Program(Under Review) | GI Genius(K211951) | Comment |
|---|---|---|---|
| Manufacturer | FUJIFILM Corporation | Cosmo Artificial Intelligence – AILtd | N/A |
| Product Code | QNP | QNP | Same |
| Intended Use /Indications forUse | This software is acomputer-assisted readingtool designed to aidendoscopists in detectingcolonic mucosal lesions(such as polyps andadenomas) in real timeduring standardendoscopy examinationsof patients undergoingscreening andsurveillance endoscopicmucosal evaluations. Thissoftware is used withstandard WLI (WhiteLight Imaging) and LCI(Linked Color Imaging)endoscopy imaging. Thissoftware is not intendedto replace clinicaldecision making. | The GI Genius System isa computer-assistedreading tool designed toaid endoscopists indetecting colonic mucosallesions (such as polypsand adenomas) in realtime during standardwhite-light endoscopyexaminations of patientsundergoing screening andsurveillance endoscopicmucosal evaluations. TheGI Genius computer-assisted detection deviceis limited for use withstandard white-lightendoscopy imaging only.This device is notintended to replaceclinical decision making. | SubstantiallyEquivalent (Theonly difference inthe intended useindications foruse between thesubject deviceand the predicatedevice is theEW10-EC02EndoscopySupport Programcan also be usedwith LCIendoscopyimaging). |
| Site | Large intestine | Large intestine | Same |
| Modality | Colonoscopy | Colonoscopy | Same |
| CAD Function | Detection | Detection | Same |
| Method of reading | Concurrent read | Concurrent read | Same |
| Algorithm(s) | The EW10-EC02 Endoscopy | The GI Genius system utilizes an | Same |
{7}------------------------------------------------
| Feature | EW10-EC02Endoscopy Support Program(Under Review) | GI Genius(K211951) | Comment |
|---|---|---|---|
| Support Program utilizes anartificial intelligence-basedalgorithm to perform the polypdetection function. | artificial intelligence-basedalgorithm to perform the polypdetection function. | ||
| Algorithm failure leading to:• False positives resulting inunnecessary patient treatment; or• False negatives resulting indelayed patient treatmentFailure to identify lesions, resultingin delayed patient treatment, due tosoftware/hardware failure including: | Algorithm failure leading to:• False positives resulting inunnecessary patient treatment; or• False negatives resulting indelayed patient treatmentFailure to identify lesions, resultingin delayed patient treatment, due tosoftware/hardware failure including: | ||
| Identified risks | • Incompatibility with hardwareand/or data source• Inadequate mapping of softwarearchitecture• Degradation of image quality• Prolonged delay of real-timeendoscopic videoFalse positive or false negative dueto user overreliance on the device | • Incompatibility with hardwareand/or data source• Inadequate mapping of softwarearchitecture• Degradation of image quality• Prolonged delay of real-timeendoscopic videoFalse positive or false negative dueto user overreliance on the device | Same |
Performance Data j.
The following testing was conducted for the EW10-EC02 Endoscopy Support Program.
• Software Verification and Validation
Software verification and validation was conducted for the EW10-EC02 Endoscopy Support Program to validate it for its intended use per the design documentation in line with recommendations outlined in General Principles of Software Validation, Guidance for Industry and FDA Staff. The EW10-EC02 Endoscopy Support Program demonstrated passing results on all applicable unit, integration, and requirements testing.
• Non-clinical Performance Testing
The purpose of the standalone performance testing is to demonstrate that the objectlevel, frame-level and overall algorithmic performance is sufficient to fulfill the indications for use of the EW10-EC02. The standalone performance evaluation was carried out using the dataset shown in Tables 2-4. This evaluation was performed in both WLI (White Light Imaging) and LCI (Linked Color Imaging) modes. Fujifilm performed this standalone performance testing using a total of 149 (WLI mode) and 144 (LCI mode) colonoscopy videos.
{8}------------------------------------------------
| Table 2: Patient demographics | ||
|---|---|---|
| Item | WLI | LCI |
| Number of Patients | 149(With lesion:119, Without lesion: 30) | 144(With lesion: 114, Without lesion: 30) |
| Age (range) | $56.6 \pm 10.8 (23 - 85)$ | $56.5 \pm 10.7 (23 - 85)$ |
| Male sex | 81 % | 81 % |
| Race | Asian 100% | Asian 100% |
| Number of Patients forscreening | 146 | 142 |
| Number of Patients forsurveillance | 3 | 2 |
hla 2: Dationt domagnaphia
Table 3: Detailed Dataset of standalone performance testing
| Totalnumber | Lesion Type | Lesion Size (mm) | Lesion Form | |||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Mode | Item | Adenoma | HP | Others | 1-5 | 6-9 | ≥ 10 | Polypoid(Type I) | Non-Polypoid(Type II) | |
| WLI | Numberof lesions | 164 | 133 | 26 | 5 | 119 | 34 | 11 | 139 | 25 |
| LCI | Numberof lesions | 154 | 127 | 25 | 2 | 112 | 32 | 10 | 131 | 23 |
Table 4: Summary of Number of Cases and Frames
| Mode | Number of Cases | Number of FramesWith lesion | Withoutlesion | Totalnumber | |||
|---|---|---|---|---|---|---|---|
| With lesion | Withoutlesion | Adenoma | HP | Others | |||
| WLI | 119 | 30 | 23,861 | 4,680 | 900 | 1,330,539 | 1,359,980 |
| 149 | 29,441 | ||||||
| LCI | 114 | 30 | 21,932 | 4,297 | 360 | 335,014 | 361,603 |
| 144 | 26,589 |
Based on the results in the case of evaluation per frame, the evaluation values for each case were calculated by counting the number of consecutive frames of each metric. Since lesion detection was performed and evaluated at the lesion level (object level) in the actual clinical cases, we set sensitivity per lesion as the primary endpoint. We evaluated whether the lower limit of the confidence interval (95%) of sensitivity per lesion of EW10-EC02, exceeds the criteria. We also evaluated FP Objects/Patient. The confidence interval (95%) was calculated by non-parametric cluster bootstrap analysis, considering within-patient correlation.
Standalone performance results
The evaluation results of object level performance for Standalone performance testing were shown in Tables 5-6 below.
| Table 5: Main evaluation results of object level performance for standalone performance testing | ||
|---|---|---|
| Item | Results | |
|---|---|---|
| Sensitivity per lesion | 95.1% | 95.5% |
{9}------------------------------------------------
| (Lesion-based sensitivity) | 91.1 - 98.3% | 91.5 - 98.7% |
|---|---|---|
| FP Objects/Patient | 1.42 | 0.76 |
| (Number of FPc per Case) | 1.09 - 1.81 | 0.42 - 1.21 |
| Target lesion | Sensitivity per Lesion | FP Objects/Patient | |||
|---|---|---|---|---|---|
| WLI mode | LCI mode | WLI mode | LCI mode | ||
| Lesion Type | Adenoma | 96.2%(92.3 - 99.3%) | 96.1%(91.9 - 99.2%) | 1.44(1.06 - 1.93) | 0.71(.26 - 1.34) |
| HP | 92.3%(80.0 - 100%) | 92.0%(79.2 - 100%) | 0.88(0.38 - 1.46) | 0.35(0.13 - 0.61) | |
| Lesion Size(mm) | 1-5 | 95.0%(89.9-99.1%) | 93.8%(88.4-98.2%) | 1.41(0.99 - 1.94) | 0.79(0.31 - 1.47) |
| 6-9 | 97.1%(90.6 -100%) | 100%(-) | 1.03(0.60 - 1.50) | 0.25(0 - 0.57) | |
| ≥10 | 90.9%(70.0 - 100%) | 100%(-) | 0.80(0.20 - 1.50) | 0(-) | |
| Lesion Form | Polypoid | 97.8%(94.9 - 100%) | 97.0%(93.4-99.3%) | 1.36(0.97 - 1.86) | 0.75(0.28 - 1.36) |
| Non-Polypoid | 80.0%(62.5 - 95.8%) | 87.0%(70.8 - 100%) | 0.83(0.38 - 1.38) | 0.18(0-0.41) | |
| Type&Size | 1-5mmAdenoma | 95.7%(90.5 - 100%) | 94.4%(88.3 - 98.9%) | 1.54(1.04 - 2.07) | 0.83(0.28 - 1.65) |
| 6-9mmAdenoma | 96.7%(89.7 - 100%) | 100%(-) | 1.19(0.69 - 1.69) | 0.24(0 - 0.60) | |
| ≥10mmAdenoma | 100%(-) | 100%(-) | 1.00(0.25 - 1.75) | 0(-) | |
| Screening | 95.0%(91.0-98.2%) | 95.4%(91.5 - 98.7%) | 1.43(1.09 - 1.82) | 0.77(0.42 - 1.25) | |
| Post-treatment surveillance | 100%(-) | 100%(-) | 0.67(0-2.0) | 0(-) | |
| Cases with Others polyps | (without any identified polypssubgroup) | 80.0%(40.0 - 100%) | 100%(-) | 0.20(0 - 0.60) | 0(-) |
| Cases without lesion | - | - | 1.80(1.07 - 2.67) | 1.03(0.47 - 1.77) |
Table 6: Evaluation results for each target lesions and cases
Figure 1 showed how detection persistence in time (the duration of time a mark persists on the same target based on an IoU overlap criterion applied to the EW10-EC02 marks across frames) correlates with sensitivity per lesion and FP Objects /Patient. This testing considers repeated marking overlays of the same target (lesions and false positives) as a single statistical event. This allows for an estimate of the number of unique targets (or objects) identified by EW10-EC02 as a function of the time those targets persist in the field of view. In this standalone performance testing, WLI has about four times more evaluation images than LCI, so there is a difference in FP Objects/Patient, but the relationship between FP Objects/Patient and Sensitivity per lesion in both modes was almost equivalent.
{10}------------------------------------------------
Image /page/10/Figure/1 description: The image contains two plots comparing sensitivity per lesion vs the number of false positives per case. The plot on the left is labeled WLI and the plot on the right is labeled LCI. Both plots show a similar trend, with sensitivity increasing rapidly as the number of false positives increases from 0 to 20, and then leveling off. The sensitivity reaches nearly 100% with a small number of false positives.
Figure 1: Figure demonstrating how detection persistence in time correlates with sensitivity per lesion and the number of FPc per "Case" (FP Objects/Patient); (L)WLI mode, (R)LCI mode
The frame-level performance is an assessment of the accuracy of the algorithm at sorting endoscopic images for quantification of false positives, false negatives, true positives, and true negatives. The image-level evaluation results were shown in the Table 7.
| Items | Results | ||
|---|---|---|---|
| WLI mode | LCI mode | ||
| Frame levelperformance | Total number of TPF | 21,166 | 21,723 |
| Total number of TNF | 1,273,229 | 317,600 | |
| Total number of FPF | 69,075 | 22,454 | |
| Total number of FNF | 8,275 | 4,866 | |
| Sensitivity perframe | Case with lesion (All lesion)(Total number of TPF/ (Total numberof TPF+ Total number of FNF)) | 71.9%% of polyps: 97.6%(71.4 - 72.4%) | 81.7%% of polyps: 99.4 %(81.2 – 82.2%) |
| Case for Screening | 72.0%(71.4 – 72.5%) | 81.8%(81.3 – 82.2%) | |
| Case for Post-treatmentsurveillance | 68.2%(64.3 – 72.0%) | 76.7%(72.2 – 81.1%) | |
| Cases with Others polyps | 73.4%(70.8 – 76.7%) | 98.6%(97.2 – 99.7%) | |
| False PositiveRate per frame | All cases(FPF / Number of all frames) | 5.08%(4.46 – 5.88%) | 6.21%(4.45 – 8.31%) |
| Case for Screening | 5.12%(4.49 – 5.92%) | 6.22%(4.54 – 8.28%) | |
| Case for Post-treatmentsurveillance | 3.43%(2.10 – 4.27%) | 4.01%(2.44 – 12.9%) | |
| Cases with Others polyps | 4.14%(3.56 – 4.83%) | 0.54%(0 – 1.07%) | |
| Cases without lesion | 5.32%(3.92 – 7.26%) | 6.42%(4.12 – 9.36%) |
Table 7: Results of frame level performance in Standalone performance testing
Figures 3-4 showed the results of ROC and FROC analysis. The AUC values of ROC curves were respectively 0.79(WLI) and 0.87(LCI). These curve and value showed that the recognizer algorithm has high accuracy.
{11}------------------------------------------------
Image /page/11/Figure/1 description: This image contains two ROC analysis graphs. The first graph is titled "ROC analysis in WLI mode (Overall view)" and has an AUC of 0.79 with a 95% confidence interval of 0.77-0.80. The second graph is titled "ROC analysis in LCI mode (Overall view)" and has an AUC of 0.87 with a 95% confidence interval of 0.86-0.88. Both graphs plot the True Positive Rate on the y-axis and the False Positive Rate on the x-axis.
Figure 3: ROC analysis; (L)WLI mode, (R)LCI mode
Image /page/11/Figure/3 description: The image contains two FROC analysis graphs, one in WLI mode and the other in LCI mode, both displaying an overall view. The x-axis of both graphs represents the number of FPc (False Positives per case), while the y-axis represents the sensitivity per lesion. The WLI mode graph shows the number of FPc per case ranging up to 200, while the LCI mode graph shows the number of FPc per case ranging up to 60. Both graphs show a similar trend, with sensitivity per lesion increasing rapidly at lower FPc values and then plateauing as FPc increases.
Figure 4: FROC analysis; (L)WLI mode, (R)LCI mode
Standalone Performance Conclusions
EW10-EC02 achieved all criteria in both modes and showed good results for each subgroup on both items of sensitivity and false positives. In addition, sensitivity analysis and FPc analysis further support robust results for EW10-EC02. ROC-AUC and FROC analysis also supports that performance of the EW10-EC02 algorithm.
Special Control Testing
- √ Pixel-level comparison of degradation of image quality due to the device; No visually detectable differences between images were found with the introduction of the EW10-EC02.
- √ Video delay due to marker annotation was 4.00 frames average (66.7 msec).
- √ Real-time endoscopic video delay due to the device was 3.67 frames average (61.2 msec).
• Human Factors
The data collected in the pivotal clinical study is sufficient to support usability as it demonstrates the ability to appropriately use the device in an actual clinical setting.
{12}------------------------------------------------
•Clinical Testing
Study Design
This study was a multi-center, prospective, randomized controlled trial. Each subject had a colonoscopy, using FUJIFILM's High Definition (HD) endoscope, video processor and the EW10-EC02 endoscopy support program. The study was conducted at 12 centers in the United States.
Subjects met all eligibility criteria were randomly allocated to 1 of 2 arms:
- (1) CAC(Computer Assisted Colonoscopy) group: inspection with computer assisted colonoscopy. Or
- (2) CC(Conventional Colonoscopy) group: inspection with conventional colonoscopy.
As per standard of care, polyps were resected when found, and sent to histopathology.
Study Population
This prospective study enrolled a total of 1,166 subjects. Of these subjects enrolled. 135 subjects were excluded from the analysis due to exclusionary reasons. A total of 1,031 subjects were analyzed, 600 were average risk subjects undergoing average risk screening colonoscopy and 565 subjects scheduled for follow-up colonoscopy due to a previous history of polyps 3 years or greater.
These subjects provided an informed consent and were aged 45 or older. Patients were not enrolled if they were pregnant, refused to give an informed consent or had a history of colon resection, Inflammatory Bowel Disease (IBD), Familial Adenomatous Polynosis (FAP). severe end-stage cardiovascular/pulmonary/liver/renal disease. A key demographics is provided below:
| Total | Randomized to | |||
|---|---|---|---|---|
| n = 1031 | CACn = 509 | CCn = 522 | P-Value | |
| Age | $59.1 \pm 9.8$ | $58.9 \pm 9.5$ | $59.3 \pm 10.1$ | 0.498 |
| Sex | 0.304 | |||
| Male | 514 (49.9%) | 262 (51.5%) | 252 (48.3%) | |
| Female | 517 (50.1%) | 247 (48.5%) | 270 (51.7%) | |
| Hispanic | 57 (5.5%) | 25 (4.9%) | 32 (6.1%) | 0.391 |
| Race | 0.448 | |||
| 1 Caucasian | 743 (72.3%) | 363 (71.6%) | 380 (73.1%) | |
| 2 Af Am | 169 (16.5%) | 86 (17.0%) | 83 (16.0%) | |
| 3 Asian | 63 (6.1%) | 37 (7.3%) | 26 (5.0%) | |
| 4 Native Haw/PI | 4 (0.4%) | 2 (0.4%) | 2 (0.4%) | |
| 5 Native Am | 2 (0.2%) | 1 (0.2%) | 1 (0.2%) | |
| 6 Other | 46 (4.5%) | 18(3.6%) | 28 (5.4%) | |
| Missing | 4 | 2 | 2 | |
| Reason for the colonoscopy | 0.116 | |||
| 1 Screening Colonoscopy | 540 (52.4%) | 254 (49.9%) | 286 (54.8%) | |
| 2 Surveillance Colonoscopy | 491 (47.6%) | 255 (50.1%) | 236 (45.2%) |
{13}------------------------------------------------
Equipment
The following Equipment was used in the study, as in real-world use of the device.
- . VP-7000 Video Processor
The Processor relays the image from the endoscope to a video monitor. The Processor provides the optional image enhancement function (BLI, LCI, FICE) at the user's option.
● BL-7000 LED Light Source
The Fujifilm endoscope employs fiber bundles to transmit light from the light source and subsequently to the body cavity. The Light Source employs four LED lamps. Brightness control is performed by the user.
● 700 series Colonoscopes
The Fujifilm 700 series Colonoscopes are intended for the visualization of the lower digestive tract, specifically for the observation, diagnosis, and endoscopic treatment of the rectum and large intestine. The endoscope is used in combination with FUJIFILM's video processors, light sources and peripheral devices such as monitor, printer, foot switch, and cart.
Study Endpoint
The purpose of this study was to demonstrate the superiority of colorectal polyp detection using computer assisted colonoscopy compared to conventional colonoscopy. To accomplish this objective, two primary endpoints were evaluated:
Co-Primary Endpoints
-
Adenoma per colonoscopy (APC)
APC, defined as the total number of histologically confirmed adenomas detected in the colonoscopy divided by the total number of colonoscopies. -
Positive predictive value (PPV)
PPV, defined as the total number of histologically confirmed adenomas detected during the colonoscopy, divided by the total number of excisions in the colonoscopy. -
Positive percent agreement (PPA)
PPA, defined as the total number of histologically confirmed Clinically Significant Excised Lesions , divided by the total number of excisions. For calculating this endpoint, clinically Significant Excised Lesions defined as follows: -
Neoplastic lesions (classical adenomas and carcinomas) י
-
Sessile serrated lesions (SSL) classified according to the serrated lesion . classification
{14}------------------------------------------------
- Hyperplastic polyps (HP) of the proximal colon (caecum, ascending colon, י hepatic flexure and transverse colon), classified according to the serrated lesion classification.
Secondary Endpoint
Additional secondary endpoints were assessed:
- Adenoma detection rate (ADR) ●
ADR, defined as the proportion of patients with at least one histologically confirmed adenomas detected in the colonoscopy.
● Sessile serrated lesion per colonoscopy (SPC)
SPC, defined as the total number of histologically confirmed sessile serrated lession detected in the colonoscopy divided by the total number of colonoscopies.
● Sessile serrated lesion detection rate (SDR)
SDR, defined as the proportion of patients with at least one histologically confirmed sessile serrated lesion detected in the colonoscopy.
● Polyp per colonoscopy (PPC)
PPC, defined as the total number of histologically confirmed clinically relevant detected in the colonoscopy divided by the total number of colonoscopies.
● Polyp detection rate (PDR)
PDR, defined as the proportion of patients with at least one histologically confirmed clinically relevant polyp detected in the colonoscopy.
-
Adverse events
Adverse Events number and severity. -
Cecal intubation rates and withdrawal times
-
Serrated Lesions per Colonoscopy (SLPC)
SLPC, defined as the number of histologically confirmed serrated detected, divided by the total number of colonoscopies.
● Serrated Lesions Detection Rate (SLDR)
SLDR, defined as the proportion of patients with at least one histologically confirmed serrated lesions detected.
● Advanced Adenoma Detection Rate (aADR)
aADR, defined as proportion of patients with at least one histologically confirmed adenoma > 10 mm or any adenoma < 10 mm, which was either of high-grade dysplasia or villous or tubulovillous
● Small Adenoma Detection Rate (sADR)
sADR, defined as proportion of patients with at least one histologically confirmed adenoma smaller than 5 mm detected
● Flat Adenoma Detection Rate (fADR)
{15}------------------------------------------------
fADR, defined as the proportion of patients with at least one histologically confirmed non-polypoid adenoma detected
- Proximal Adenoma Detection Rate (pADR)
pADR, defined as the proportion of patients with at least one histologically confirmed adenoma in transverse colon, Hepatic F, ascending colon, or the cecum
● False Positive Rate (FPR)
FPR, defined as the proportion of colorectal lesions resected or biopsied and subsequently not histologically confirmed to be clinically relevant colorectal polyps. All the biopsied or ablated specimens, which were histologically confirmed not to be polyps (e.g., normal mucosa, inflammatory tissue, stool, or debris, etc.), were classified as False Positive.
● True Histology Rate (THR)
THR, defined as Total number of histologically confirmed adenomas, sessile serrated lesions, and large (>10 mm) hyperplastic polyps of the proximal colon resected, divided by the total number of excisions in the colonoscopy.
Study Primary Results
The study met primary success criteria with APC in CAC being superior with a p value of 0.018. PPV criteria was also met with a margin of -9.56%.
Adenoma Per Colonoscopy: APC
| Totaln = 1031 | Randomized to | Incidence RateRatio(95% CI) | Difference in APC(95% CI) | P-Value | ||
|---|---|---|---|---|---|---|
| CACn = 509 | CCn = 522 | |||||
| Mean Number ofAdenomas (+/- sd) | 0.919 ± 1.548 | 0.990 ± 1.610 | 0.849 ± 1.484 | 1.17 (1.01, 1.36) | 0.141 (0.01, 0.28) | 0.018 |
Positive Predictive Value: PPV
| Total | Randomized to | |||
|---|---|---|---|---|
| n = 1857 | CACn = 1037 | CCn = 820 | Bootstrapped95% Confidence Interval | |
| Adenoma Detected | 947 (51.0%) | 504 (48.6%) | 443 (54.0%) | -9.56%, -1.48% |
Positive Percent Agreement (PPA)
| Total | Randomized to | Bootstrapped(95% CI) | ||
|---|---|---|---|---|
| n = 1857 | CACn = 1037 | CCn = 820 | ||
| Positive Detected | 1172 (63.1%) | 629 (60.7%) | 543 (66.2%) | -10.50%, -2.30% |
Study Secondary Results
Additional secondary endpoints of note were:
- . Polyp per colonoscopy (PPC) with a p-value <0.001, defined as any neoplastic or hyperplastic polyp.
{16}------------------------------------------------
- . Serrated Lesions per Colonoscopy (SLPC) and Serrated Lesions Detection Rate (SLDR) with p-values of <0.001 and 0.027 respectively.
- . Proximal Adenoma Detection Rate (pADR), just missing statistical significance with a p-value of 0.053.
- . False Positive Rate (FPR) was non-inferior with a 95% confidence interval of 1.39 - 7.82%.
All other secondary endpoints did not meet criteria for statistical significance. No significant differences were noted in adverse events and withdrawal times.
Poolability and interaction analysis showed statistically significant interaction (p-value = 0.019) when assessing Screening versus Surveillance colonoscopy. Surveillance colonoscopies had a significant affect on APC. PPV and ADR was not significant and similar between the 2 strata.
Adenoma Detection Rate (ADR)
| Total | Randomized to | ||||
|---|---|---|---|---|---|
| n = 1031 | CACn = 509 | CCn = 522 | Absolute Difference(95% CI) | P-Value | |
| Any adenomaDetected | 462 (44.8%) | 238 (46.8%) | 224 (42.9%) | 3.85% (2.22%,9.91%) | 0.214 |
Sessile Serrated Lesion Per Colonoscopy (SPC)
| Total | Randomized to | ||||
|---|---|---|---|---|---|
| n = 1031 | CACn = 509 | CCn = 522 | Incidence Rate Ratio(95% CI) | P-Value | |
| Mean SessileSerrated | 0.150 ± 0.490 | 0.171 ± 0.502 | 0.130 ± 0.478 | 1.31 (0.96, 1.80) | 0.094 |
Sessile Serrated Lesion Detection Rate (SDR)
| Total | Randomized to | ||||
|---|---|---|---|---|---|
| n = 1031 | CACn = 509 | CCn = 522 | Absolute Difference(95% CI) | P-Value | |
| Any SSDetected | 119 (11.5%) | 66 (13.0%) | 53 (10.2%) | 2.81% (-1.09%,6.72%) | 0.157 |
Polyp Per Colonoscopy (PPC)
| Total | Randomized to | Incidence Rate Ratio(95% CI) | P-Value | ||
|---|---|---|---|---|---|
| n = 1031 | CACn = 509 | CCn = 522 | |||
| Mean Numberof Polyps | 1.501 ± 1.941 | 1.680 ± 2.070 | 1.328 ± 1.791 | 1.27 (1.14, 1.40) | <0.001 |
Polyp Detection Rate (PDR)
| Total | Randomized to | ||||
|---|---|---|---|---|---|
| n = 1031 | CACn = 509 | CCn = 522 | Absolute Difference(95% CI) | P-Value | |
| Any PolypDetected | 635 (61.6%) | 325 (63.9%) | 310 (59.4%) | 4.46% (-1.47%,10.39%) | 0.140 |
{17}------------------------------------------------
Adverse Events
| Total | Randomized to | Absolute Difference(95% CI) | P-Value | ||
|---|---|---|---|---|---|
| n = 1031 | CACn = 509 | CCn = 522 | |||
| Adverse EventsYesNo | 8 (0.8%)1023 (99.2%) | 5 (1.0%)504 (99.0%) | 3 (0.6%)519 (99.4%) | 0.41% (-0.67%,1.48%) | 0.501 |
Cecal Intubation
| Total | Randomized to | P-Value | ||
|---|---|---|---|---|
| n = 1031 | CACn = 509 | CCn = 522 | ||
| Cecal Intubation | 1033 (100.0%) | 510 (100.0%) | 523 (100.0%) | NA |
Withdrawal Times
| Total | Randomized to | P-Value | ||
|---|---|---|---|---|
| CAC | CC | |||
| n = 1031 | n = 509 | n = 522 | ||
| WD Time in Seconds | $663.2 \pm 282.4$ | $677.5 \pm 275.3$ | $649.3 \pm 288.8$ | 0.109 |
Serrated Lesions Per Colonoscopy (SLPC)
| Total | Randomized to | ||||
|---|---|---|---|---|---|
| n = 1031 | CACn = 509 | CCn = 522 | Incidence Rate Ratio(95% CI) | P-Value | |
| Mean Serrated | 0.581 ± 1.142 | 0.690 ± 1.249 | 0.475 ± 1.016 | 1.45 (1.23, 1.71) | <0.001 |
Serrated Lesions Detection Rate (SLDR)
| Total | Randomized to | ||||
|---|---|---|---|---|---|
| n = 1031 | CACn = 509 | CCn = 522 | Absolute Difference(95% CI) | P-Value | |
| Any Serrated Detected | 333 (32.3%) | 181 (35.6%) | 152 (29.1%) | 6.44% (0.74%,12.14%) | 0.027 |
Advanced Adenoma Detection Rate (aADR)
| Total | Randomized to | ||||
|---|---|---|---|---|---|
| n = 1031 | CACn = 509 | CCn = 522 | Absolute Difference(95% CI) | P-Value | |
| Any AdvancedAdenoma Detected | 66 (6.4%) | 33 (6.5%) | 33 (6.3%) | 0.16% (-2.83%,3.15%) | 0.915 |
Small Adenoma Detection Rate (sADR)
| Total | Randomized to | Absolute Difference(95% CI) | P-Value | ||
|---|---|---|---|---|---|
| n = 1031 | CACn = 509 | CCn = 522 | 5.1% (-0.63%,10.83%) | 0.081 | |
| Any Small AdenomaDetected | 340 (33.0%) | 181 (35.6%) | 159 (30.5%) |
Flat Adenoma Detection Rate (fADR)
{18}------------------------------------------------
| Total | Randomized to | Absolute Difference(95% CI) | P-Value | ||
|---|---|---|---|---|---|
| n = 1031 | CACn = 509 | CCn = 522 | |||
| Any fADR Detected | 64 (6.2%) | 34 (6.7%) | 30 (5.7%) | 0.93% (-2.02%,3.88%) | 0.534 |
Proximal Adenoma Detection Rate (pADR)
| Total | Randomized to | ||||
|---|---|---|---|---|---|
| n = 1031 | CACn = 509 | CCn = 522 | Absolute Difference(95% CI) | P-Value | |
| Any pADR Detected | 357 (34.6%) | 191 (37.5%) | 166 (31.8%) | 5.72% (-0.08%,11.53%) | 0.053 |
False Positive Rate (FPR)
| Total | Randomized to | Bootstrapped95% confidence interval | ||
|---|---|---|---|---|
| n = 1857 | CACn = 1037 | CCn = 820 | ||
| FPR Detected | 305 (16.4%) | 182 (17.6%) | 123 (15.0%) | 1.39%, 7.82% |
True Histology Rate (THR)
| Total | Randomized to | Bootstrapped95% confidence interval | ||
|---|---|---|---|---|
| n = 1857 | CACn = 1037 | CCn = 820 | ||
| THR Detected | 1102 (59.3%) | 591 (57.0%) | 511 (62.3%) | -10.3%, -2.06% |
Interaction with Screening/Surveillance
| STRATA | Total | Randomized to | Incidence RateRatio(CAC vs CC,95% CI) | P-Value | InteractionP-Value | ||
|---|---|---|---|---|---|---|---|
| CAC | CC | ||||||
| APC | |||||||
| Screening | Mean Numberof Adenomas(+/- sd) | 0.763 ± 1.491 | 0.748 ± 1.322 | 0.776 ± 1.628 | 0.96 (0.79, 1.17) | 0.862 | 0.019 |
| Surveillance | Mean Numberof Adenomas(+/- sd) | 1.090 ± 1.593 | 1.231 ± 1.824 | 0.936 ± 1.285 | 1.31(1.11, 1.56) | 0.002 | |
| PPV | |||||||
| Screening | AdenomaDetected | 412 (47.2%) | 190 (43.5%) | 222 (51.0%) | 0.892 | ||
| Surveillance | AdenomaDetected | 535 (54.3%) | 314 (52.3%) | 221 (57.4%) | |||
| ADR | |||||||
| Screening | ADR | 209 (38.7%) | 101 (39.8%) | 108 (37.8%) | 0.696 | ||
| Surveillance | ADR | 253 (51.5%) | 137 (53.7%) | 116 (49.2%) |
Polyp Characteristics
Polyps were classified according to their size, location and morphology (pedunculated, sessile and non-polypoid). Non-polypoid (flat and depressed) lesions were defined as lesions endoscopically high less than half wide, according to Paris classification. Location was considered proximal if proximal to the splenic flexure. On the basis of
{19}------------------------------------------------
histological examination, polyps were categorized according to revised Vienna classification and serrated lesion classification.
| Total | Randomized to | P-Value | |||
|---|---|---|---|---|---|
| CAC | CC | ||||
| n = 1857 | n = 1037 | n = 820 | |||
| Polyp Location | 0.070 | ||||
| Cecum | 160 (8.7%) | 98 (9.5%) | 62 (7.6%) | ||
| Ascending | 342 (18.5%) | 197 (19.1%) | 145 (17.8%) | ||
| Hepatic F | 61 (3.3%) | 22 (2.1%) | 39 (4.8%) | ||
| Transverse | 422 (22.8%) | 234 (22.7%) | 188 (23.1%) | ||
| Splenic F | 18 (1.0%) | 8 (0.8%) | 10 (1.2%) | ||
| Descending | 243 (13.1%) | 136 (13.2%) | 107 (13.1%) | ||
| Sigmoid | 409 (22.1%) | 228 (22.1%) | 181 (22.2%) | ||
| Rectum | 193 (10.4%) | 110 (10.6%) | 83 (10.2%) | ||
| Missing | 9 | 4 | 5 | ||
| Polyp Shape | 0.002 | ||||
| lp | 96 (5.2%) | 41 (4.0%) | 55 (6.7%) | ||
| ls | 1502 (80.9%) | 868 (83.7%) | 634 (77.3%) | ||
| lla | 224 (12.1%) | 109 (10.5%) | 115 (14.0%) | ||
| llb | 19 (1.0%) | 12 (1.2%) | 7 (0.9%) | ||
| llc | 4 (0.2%) | 3 (0.3%) | 1 (0.1%) | ||
| III | 1 (0.1%) | 1 (0.1%) | 0 (0.0%) | ||
| Other | 11 (0.6%) | 3 (0.3%) | 8 (1.0%) | ||
| Polyp Size (mm) | 0.003 | ||||
| 1.0 to <5 | 1182 (63.9%) | 692 (66.9%) | 490 (60.0%) | ||
| 5 to <10 | 556 (30.1%) | 291 (28.1%) | 265 (32.5%) | ||
| 10 to 60.0 | 112 (6.1%) | 51 (4.9%) | 61 (7.5%) | ||
| Missing | 7 | 3 | 4 | ||
| Non-neoplastic | 0.025 | ||||
| Not NONNEOPLASTIC | 1111 (59.8%) | 592 (57.1%) | 519 (63.3%) | ||
| Hyperplastic | 441 (23.7%) | 263 (25.4%) | 178 (21.7%) | ||
| Other | 305 (16.4%) | 182 (17.6%) | 123 (15.0%) | ||
| Neoplastic | 0.014 | ||||
| NOT NEOPLASTIC | 751 (40.4%) | 445 (42.9%) | 306 (37.3%) | ||
| Adenoma | 927 (49.9%) | 495 (47.7%) | 432 (52.7%) | ||
| Traditional Serrated | 3 (0.2%) | 1 (0.1%) | 2 (0.2%) | ||
| Sessile Serrated | 152 (8.2%) | 86 (8.3%) | 66 (8.0%) | ||
| Villous | 16 (0.9%) | 5 (0.5%) | 11 (1.3%) | ||
| High-grade dysplasia | 4 (0.2%) | 4 (0.4%) | 0 (0.0%) | ||
| Submucosal Invasion | 3 (0.2%) | 1 (0.1%) | 2 (0.2%) | ||
| Other | 1 (0.1%) | 0 (0.0%) | 1 (0.1%) |
Clinical Testing Conclusions
With meeting primary success criteria and demonstrating additional secondary benefits, the Computer Assisted AI Colonoscopy (CAC) is an appropriate aid to endoscopists, trained in intestinal polyp detection, to further assist the clinician through detection of suspected findings during the exam, as a video image superimposed on the endoscope monitor.
- k. Conclusion
{20}------------------------------------------------
The EW10-EC02 Endoscopy Support Program has the same intended uses and similar indications, technological characteristics, and principles of operation as its predicate device, GI Genius (K211951). The differences in indications and technological characteristics between the subject and predicate devices do not raise new concerns regarding safety and effectiveness as demonstrated by the non-clinical and clinical performance evaluation results. Therefore, the EW10-EC02 Endoscopy Support Program can be considered substantially equivalent to the similar legally marketed device.
§ 876.1520 Gastrointestinal lesion software detection system.
(a)
Identification. A gastrointestinal lesion software detection system is a computer-assisted detection device used in conjunction with endoscopy for the detection of abnormal lesions in the gastrointestinal tract. This device with advanced software algorithms brings attention to images to aid in the detection of lesions. The device may contain hardware to support interfacing with an endoscope.(b)
Classification. Class II (special controls). The special controls for this device are:(1) Clinical performance testing must demonstrate that the device performs as intended under anticipated conditions of use, including detection of gastrointestinal lesions and evaluation of all adverse events.
(2) Non-clinical performance testing must demonstrate that the device performs as intended under anticipated conditions of use. Testing must include:
(i) Standalone algorithm performance testing;
(ii) Pixel-level comparison of degradation of image quality due to the device;
(iii) Assessment of video delay due to marker annotation; and
(iv) Assessment of real-time endoscopic video delay due to the device.
(3) Usability assessment must demonstrate that the intended user(s) can safely and correctly use the device.
(4) Performance data must demonstrate electromagnetic compatibility and electrical safety, mechanical safety, and thermal safety testing for any hardware components of the device.
(5) Software verification, validation, and hazard analysis must be provided. Software description must include a detailed, technical description including the impact of any software and hardware on the device's functions, the associated capabilities and limitations of each part, the associated inputs and outputs, mapping of the software architecture, and a description of the video signal pipeline.
(6) Labeling must include:
(i) Instructions for use, including a detailed description of the device and compatibility information;
(ii) Warnings to avoid overreliance on the device, that the device is not intended to be used for diagnosis or characterization of lesions, and that the device does not replace clinical decision making;
(iii) A summary of the clinical performance testing conducted with the device, including detailed definitions of the study endpoints and statistical confidence intervals; and
(iv) A summary of the standalone performance testing and associated statistical analysis.