(43 days)
The monitors are intended to be used for monitoring, storing, and to generate alarms for, multiple physiological parameters of adults and pediatrics. The monitors are intended for use by trained healthcare professionals in hospital environments.
The LM-8 monitor monitors parameters such as ECG (3-lead, 12-lead selectable), Respiration (RESP), Functional arterial oxygen saturation (SpO2), Invasive or noninvasive blood pressure (dual-IBP, NIBP), Temperature (dual-TEMP), Expired CO2 and Quick Temperature (Quick TEMP).
The LM-10 monitor monitors parameters such as ECG (3-lead, 12-lead selectable), Respiration (RESP), Functional arterial oxygen saturation (SpO2), Invasive or noninvasive blood pressure (2/4 channels IBP, NIBP), Cardiac Output (C.O.), Temperature (dual-TEMP) and Expired CO2.
The LM-12 monitor monitors parameters such as ECG (3-lead, 12-lead selectable), Respiration (RESP), Functional arterial oxygen saturation (SpO2), Invasive or noninvasive blood pressure (2/4 chamels IBP, NIBP), Cardiac Output (C.O.), Temperature (dual-TEMP), Expired CO2 and Anesthetic gas (AG).
The LM-15 monitors parameters such as ECG (3-lead, 5-lead, 12-lead selectable), Respiration (RESP), Functional arterial oxygen saturation (SpO2), Invasive or noninvasive blood pressure (2/4 channels IBP, NIBP), Cardiac Output (C.O.), Temperature (dual-TEMP), Expired CO2 and Anesthetic gas (AG).
The arrhythmia detection and ST Segment analysis are intended for adult and pediatric patients.
The monitors are not intended for MRI environments.
LM-8, LM-10, LM-12 and LM-15, patient monitor integrates parameter measuring modules, display and recorder in one device, featuring in compactness, lightweight and portability. Replaceable built-in battery facilitates patient transport. Large high-resolution display provides clear view of 10 waveforms and full monitoring parameters. Patient Monitor can monitor vital signal such as ECG, respiration (RESP), non-invasive blood pressure (NIBP), oxygen saturation of the blood (SpO2), temperature (TEMP), invasive blood pressure (IBP), cardiac output (C.O.), CO2 and anesthetic gas (AG). Those signals are digitized, processed and examined for alarm conditions, after that presents all those information on the color TFT display. The monitor also provides advantageous operating control for the user.
The provided text is an FDA 510(k) summary for a Patient Monitor (models LM-8, LM-10, LM-12, LM-15). It primarily focuses on demonstrating substantial equivalence to predicate devices through technical comparisons and compliance with general performance standards.
However, the document does not contain the detailed information necessary to answer all aspects of your request regarding acceptance criteria and a study proving the device meets those criteria in the context of an AI/algorithm-driven medical device performance study.
Specifically, the document lacks:
- A explicit table of acceptance criteria for algorithm performance (e.g., sensitivity, specificity, F1-score for arrhythmia detection).
- Detailed results of a study demonstrating the device meets specific performance criteria for arrhythmia detection or ST-segment analysis (beyond basic functional checks).
- Information on sample size for test sets directly related to algorithm performance (as opposed to overall device safety/functionality).
- Data provenance, number of experts for ground truth, adjudication methods, or MRMC studies, which are typical for AI/ML device evaluations.
- Training set details for any AI/ML components.
The "Performance data" section refers to "Clinical data" for validation, but these appear to be general functional validation tests on physiological parameters (ECG, RESP, SpO2, NIBP, etc.) to ensure the monitors function as intended, rather than a specific study to validate the performance of the arrhythmia detection and ST Segment analysis algorithm against clinical ground truth. The statement "The clinical data demonstrate that the subject devices perform comparably to the predicate device that is currently marketed for the same intended use" is a high-level conclusion without supporting details beyond the comparative features table.
Based on the provided text, here's what can be extracted and what is missing:
1. A table of acceptance criteria and the reported device performance:
The document doesn't provide a specific table of quantitative acceptance criteria for the arrhythmia detection and ST segment analysis algorithm (e.g., sensitivity, specificity thresholds) and corresponding reported performance metrics. It lists general parameters and their measurement ranges, which are functional specifications, not performance criteria for an arrhythmia detection algorithm.
General device functional specifications (from comparison table, not acceptance criteria for algorithm):
| Parameter | Proposed Device (LM-8, LM-10, LM-12, LM-15) | Predicate Devices (K113623, K131971) | Reported Performance |
|---|---|---|---|
| ECG Parameters | |||
| Lead Mode | 3, 5, 12 Electrodes (depending on model) | 3, 5, 12 Electrodes (depending on model) | Same as predicate |
| Arrhythmia analyses | ASYSTOLE, VFIB/VTAC, COUPLET, VT> 2, BIGEMINY, TRIGEMINY, VENT, R on T, PVC, TACHY, BRADY, MISSED BEATS, IRR, VBRADY, PNC, PNP | Same | Same as predicate (implies detection of these) |
| ST value Measurement Range | -2.0 mV to +2.0 mV | Same | Same as predicate |
| HR Measurement range | ADU: 15-300 bpm, PED/NEO: 15-350 bpm | Same | Same as predicate |
| PVC Range | ADU: (0 to 300) PVCs/min, PED/NEO: (0 to 350) PVCs/min | Not explicitly stated for predicate in comparison table, but "Same" is indicated. | Implied performance within these ranges. |
| RESP Module | |||
| Principle of Operation | Impedance between RA-LL, RA-LA | Same | Same as predicate |
| Measurement Range | Adult: 0-120 rpm, Pediatric/neonate: 0-150 rpm | Same | Same as predicate |
| NIBP Module | |||
| Technique | Oscillometry | Same | Same as predicate |
| Measurement Range | Adult/Pediatric/Neonatal Systolic, Diastolic, Mean ranges provided (e.g., Adult Systolic 40-270) | Same | Same as predicate |
| PR from NIBP Measurement range | 40-240 bpm | Same | Same as predicate |
| SpO2 Module | |||
| Measurement Range | SpO2: 0-100%, Pulse Rate: 25-300 bpm | Same | Same as predicate |
| Temperature Module | |||
| Number of channels | 2 | Same | Same as predicate |
| Measurement Range | 0°C to 50°C (32°F to 122°F) | Same | Same as predicate |
| IBP Module | |||
| Measurement Range | PA/PAWP, CVP/RAP/LAP/ICP, P1/P2 ranges (e.g., P1/P2: -50 to +300 mmHg) | Same | Same as predicate |
| C.O. Module | |||
| Technique | Thermodilution Technique | Same | Same as predicate |
| Measurement Range | C.O.: 0.1-20L/min, TB: 23-43°C, TI: -1-27°C | Same | Same as predicate |
| CO2 Module | |||
| Measure Parameters | EtCO2, FiCO2, AwRR | Same | Same as predicate |
| Measuring Range | CO2: 0-150mmHg, AwRR: 2-150rpm | Same | Same as predicate |
| AG Module | |||
| Measure Parameters | CO2, N2O, O2, HAL, ISO, ENF, SEV, DES, awRR, MAC. | Same | Same as predicate |
| Measuring Range | Ranges for each gas and AwRR (e.g., CO2: 0-25 vol%) | Same | Same as predicate |
The document notes that "The arrhythmia detection and ST Segment analysis are intended for adult and pediatric patients" and that "Clinical tests were performed on the LM-8, LM-12 and LM-15 monitors to validate their performance in terms of ECG...". However, it does not specify what constituted "validation" for these particular algorithmic features or what the performance metrics were. The "Conclusion" states: "The clinical data demonstrate that the subject devices perform comparably to the predicate device that is currently marketed for the same intended use." This is the reported device performance for these features: "comparable to predicate."
2. Sample sized used for the test set and the data provenance:
- Sample Size for Test Set: Not specified for the "Clinical data" related to ECG/arrhythmia/ST validation. The statement is general: "Clinical tests were performed on the LM-8, LM-12 and LM-15 monitors..."
- Data Provenance: Not specified (e.g., country of origin, retrospective or prospective).
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g. radiologist with 10 years of experience):
- Not specified. The document does not describe how ground truth for arrhythmia or ST segment analysis was established for clinical testing.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set:
- Not specified.
5. If a multi reader multi case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance:
- No, an MRMC study is not mentioned. This device is a monitor, not an AI-assisted diagnostic tool for interpretation by a human reader in the typical sense of an MRMC study for imaging. It provides "arrhythmia detection and ST Segment analysis" algorithms directly to the user.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- The "Clinical data" statement indicates the monitors' performance was validated. Given the nature of a patient monitor, the arrhythmia and ST segment analysis would inherently be "standalone" algorithmic functions integrated into the device, providing automated analysis. However, specific performance metrics (like sensitivity/specificity of the algorithm itself) from this standalone evaluation are not presented.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc):
- Not specified. It is generally implied that such devices are validated against accepted physiological measurement standards and potentially manually confirmed ECG interpretations, but the document does not detail this for the arrhythmia/ST segment analysis.
8. The sample size for the training set:
- Not applicable/Not specified. The document does not indicate that the arrhythmia detection or ST segment analysis algorithms utilize machine learning or require a "training set" in the sense of AI/ML development. It's likely these are based on established rule-based or signal processing algorithms, not learned from data.
9. How the ground truth for the training set was established:
- Not applicable. (See #8)
In summary, this 510(k) submission successfully demonstrates substantial equivalence through technical specifications, comparisons to predicate devices, and compliance with general safety and performance standards (e.g., electrical safety, EMC, biocompatibility, software verification/validation). However, it does not detail a specific performance study for its arrhythmia detection and ST segment analysis algorithms in a way that typically applies to AI/ML clearance, which would include explicit acceptance criteria, detailed test set characteristics, and ground truth methodologies. The "clinical data" section is very high-level and only states comparability to predicate devices.
§ 870.1025 Arrhythmia detector and alarm (including ST-segment measurement and alarm).
(a)
Identification. The arrhythmia detector and alarm device monitors an electrocardiogram and is designed to produce a visible or audible signal or alarm when atrial or ventricular arrhythmia, such as premature contraction or ventricular fibrillation, occurs.(b)
Classification. Class II (special controls). The guidance document entitled “Class II Special Controls Guidance Document: Arrhythmia Detector and Alarm” will serve as the special control. See § 870.1 for the availability of this guidance document.