A patient registry of 201 healthy Caucasian subjects was used to construct a normative database for RNFL thickness measurements, allowing for quantitative comparison of patient results against a normal population.
201 healthy Caucasian subjects (111 male, 90 female; age range 18-78 years); Sample Size: 201
Not applicable for this study
RNFL thickness measurements
AI Performance
Output
Algorithm
Acceptance
Observed
Dev DS
Dev Readers
Test DS
Test Readers
Retinal Nerve Fiber Layer (RNFL) thickness
—
—
Coefficients of variation within specified range for reproducibility and repeatability.
201 healthy Caucasian subjects (111 male, 90 female, mean age 48.2 ± 14.5 years).
—
Agreement study comparing Spectralis to Stratus OCT in 101 healthy subjects and 183 glaucoma patients.
—
Indications for Use
The Spectralis HRA+OCT is a non-contact ophthalmic diagnostic imaging device. It is intended for viewing the posterior segment of the eye, including two- and three-dimensional imaging, cross-sectional imaging, fundus photography, and fluorescence imaging (fluorescein, indocyanine green and autofluorescence), and to perform measurements of ocular anatomy and ocular lesions. The device is indicated as an aid in the detection and management of various ocular diseases including: age-related macular degeneration, macular edema, diabetic retinopathy, retinal and choroidal vascular diseases, glaucoma, and for viewing geographic atrophy as well as changes in the eye that result from neurodegenerative diseases. The Spectralis HRA+OCT includes a retinal nerve fiber layer thickness normative database, which is used to quantitatively compare the retinal nerve fiber layer in the human retina to a database of Caucasian normal subjects; the classification result is valid only for Caucasian subjects.
Device Story
Spectralis HRA+OCT combines optical coherence tomography (OCT) and confocal scanning laser ophthalmoscopy (cSLO) for posterior segment imaging. Inputs: simultaneous OCT cross-sectional scans and cSLO reflectance/angiography/autofluorescence images. Operation: laser scanning camera captures data; host computer processes and displays images side-by-side. Clinical use: ophthalmologists/clinicians use device in clinics to visualize ocular structures and measure retinal nerve fiber layer (RNFL) thickness. Output: high-resolution images and quantitative RNFL measurements compared against an age-adjusted normative database. Clinical decision-making: aids in diagnosing and monitoring progression of retinal diseases and glaucoma. Patient benefit: non-contact, high-resolution visualization and objective quantification of ocular pathology.
Clinical Evidence
Clinical evaluation included reproducibility and repeatability studies in human volunteers, confirming coefficients of variation within specified ranges. Normative database established using 201 healthy Caucasian subjects (ages 18-78). Agreement study (n=101 healthy, n=183 glaucoma) compared Spectralis to Stratus OCT; results showed good linear correlation but indicated measurements are not interchangeable. Case reports confirmed no artifacts and predictable RNFL decrease in glaucoma patients.
Technological Characteristics
Non-contact ophthalmic imaging system; combines OCT and cSLO. Laser scanning camera, camera mount, operation panel, power supply, host PC. Class 1 laser product (IEC 60825-1). Complies with IEC 60601-1, 60601-1-2, 60601-1-4. Software-based RNFL normative database (Caucasian only).
Indications for Use
Indicated for patients requiring posterior segment imaging and assessment of ocular anatomy/lesions. Aids in detection/management of age-related macular degeneration, macular edema, diabetic retinopathy, retinal/choroidal vascular diseases, glaucoma, geographic atrophy, and neurodegenerative eye changes. RNFL normative database valid only for Caucasian subjects.
Regulatory Classification
Identification
An ophthalmoscope is an AC-powered or battery-powered device containing illumination and viewing optics intended to examine the media (cornea, aqueous, lens, and vitreous) and the retina of the eye.
Special Controls
*Classification.* Class II (special controls). The device, when it is an AC-powered opthalmoscope, a battery-powered opthalmoscope, or a hand-held ophthalmoscope replacement battery, is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 886.9.
{0}------------------------------------------------
# 510(k) Summary
## Manufacturer and Submitter
| Company Name: | Heidelberg Engineering GmbH | |
|------------------------|-----------------------------------------------------------------------------------------|----|
| Company Address: | Tiergartenstrasse 15 | |
| | 69121 Heidelberg, Germany<br>phone: +49 / 6221 / 64 643 0<br>fax: +49 / 6221 / 64 63 62 | 00 |
| Contact Person: | Dr. Gerhard Zinser | |
| Date Summary Prepared: | September 27, 2010 | |
1 1 2010
## Device
| Trade/Device Name: | Spectralis™ HRA+OCT |
|-----------------------|-------------------------------------------------|
| Common/Usual Name: | Retina Angiograph / Optical Coherence Tomograph |
| Classification Name: | Ophthalmoscope, AC-powered |
| Regulation Number: | 21 CFR 886.1570 |
| Product Code: | MYC, OBO |
| Classification Panel: | Ophthalmic |
| Classification: | Class II |
### Device Description
The Spectralis HRA+OCT is a real-time imaging system of the posterior segment of the human eye and for aiding in the assessment and management of various diseases of the posterior segment, such as age-related macular degeneration, diabetic retinopathy, and glaucoma. The device is a combination of optical coherence tomography (OCT) with confocal scanning laser ophthalmoscopy (cSLO). OCT imaging includes high-resolution cross-sectional imaging of ocular structures (e.g., retina, macula, optic nerve head); cSLO imaging includes high-resolution and dynamic infrared reflectance, blue reflectance, fluorescein angiography, indocyanine green angiography, and autofluorescence imaging. OCT images and cSLO images are acquired simultaneously and are viewed side-by-side on the computer screen. Images are acquired and stored using Spectralis operation software, which runs on a standard personal computer. Spectralis components include a laser scanning camera, camera mount with headrest, operation panel, power supply box, operation software, and host computer.
#### Intended Use/Indications for Use
The Spectralis HRA+OCT is a non-contact ophthalmic diagnostic imaging device. It is intended for viewing the posterior segment of the eye, including two- and three-dimensional imaging, cross-sectional imaging, fundus photography, and fluorescence imaging (fluorescein, indocyanine green and autofluorescence), and to perform measurements of ocular anatomy and ocular lesions. The device is indicated as an aid in the detection and management of various ocular diseases including: age-related macular degeneration, macular edema, diabetic retinopathy, retinal and choroidal vascular diseases, glaucoma, and for viewing geographic atrophy as well as changes in the eye that result from neurodegenerative diseases. The Spectralis HRA+OCT includes a retinal nerve fiber layer thickness normative database, which is used to quantitatively compare the retinal nerve fiber layer in the human retina to a database of Caucasian normal subjects; the classification result is valid only for Caucasian subjects.
l
{1}------------------------------------------------
# Substantial Equivalence Analysis
The Spectralis HRA+OCT has the same intended use and basic technological characteristics as the Heidelberg HRA2/OCT (K063191). The device has been updated to include new optical layout and components, new electronic components, new mechanical design, and new software including the addition of a normative database for RNFL thickness. These differences do not adversely affect safety or effectiveness, as verified with clinical and bench testing. Predicate devices incorporating similar technology features to the upgraded Spectralis device include: the STATUS OCT with Retinal Nerve Fiber Layer Normative Database (K030433) for the similarity in having a normative database, and the IRI Integrated Retinal Imager (K062295) for the autofluorescence intended use.
# Performance - Bench Testing
The Spectralis HRA+OCT has been tested according to IEC 60601-1, IEC 60601-1-2, and IEC 60601-1-4 and was found to meet all requirements. The system is a laser product of Class 1 according to 21 CFR §1040.10 and complies with IEC 60825-1.
Accuracy testing was performed to determine the accuracy of measurements in the OCT images, as well as in the reflectance and angiography images. Results of this study confirmed accuracy of measured values compared to one another and compared to the true value, verifying that the performance of the Spectralis device is accurate and within stated specifications.
# Performance - Clinical Evaluation
## Reproducibility and Repeatability
A study in human volunteers was conducted to ensure Repeatability and Reproducibility of Spectralis HRA+OCT device measurements. All possible measures were evaluated for the Spectralis HRA+OCT. For reproducibility and repeatability, the coefficients of variation of the measured endpoints were within the specified range for this device.
### Normative Database
The normative database includes 201 subjects of Caucasian origin enrolled in a patient registry. Demographically, the subjects included: 111 male, 90 female, mean (SD) age 48.2 ± 14.5 years, age range 18 to 78 years. Included subjects had no history of glaucoma, normal intraocular pressure, normal visual field, normal appearance of optic disc, etc. Screening for entry into the study included patient history and physical examination to determine if eyes were "normal" by two ophthalmologists.
If subjects had a normal examination and no medical history of ocular pathology, images were acquired and measurements were then made using the Spectralis HRA+OCT device per Instructions for Use to measure RNFL thickness. Results were compiled to form a normative database of RNFL thickness. The database is limited by a sample size of 201 subjects (1 case in the <20 years age group, 13 cases in the>70 years age group), Caucasian ethnicity, and inclusion of subjects with refractive errors in the range from +5 diopters to -7 diopters.
Normative database limitations described above should be considered when examining subjects whose characteristics differ from those included in the database.
{2}------------------------------------------------
RNFL thickness measurements made with the Spectralis HRA +OCT are compared to the normative database. For that purpose, an age-adjustment of the normal database is performed and percentile limits of the normal distribution are computed as follows.
For each subject in the normative database, RNFL thickness at each point along the circle scan, mean RNFL thickness along the whole circle scan (global), and mean RNFL thickness within certain sectors of the circle scan (temporal, temporal-inferior, temporal-inferior, nasal, nasalsuperior, nasal-inferior) were measured. From these measurements, the age-adjusted percentiles of the database sample of 201 healthy Caucasians were determined. These age-adjusted percentiles form the basis for highlighting a result as being within or outside the normal limits (greater than the 5th or less than the 1st percentile of the database sample). The meaning of the nth percentile is that n percent of normal Caucasian subjects from the database sample have a RNFL thickness of less than or equal to this value.
Age-adjustment of the normal data is based on the linear regression of RNFL thickness vs. age of the normal subjects. Global RNFL thickness and RNFL thickness in the sectors temporal, temporal-superior, temporal-inferior, and nasal-inferior showed negative slopes (decrease of RNFL thickness) with age in the normal data and are adjusted. RNFL thicknesses in the sectors nasal and nasal-superior showed insignificant positive slopes with age in the normal data and are not adjusted.
As an example for the effect of age, the following tables show the values of the 1st and 5th percentile of RNFL thickness global and in the six sectors for ages 45 and 65 years. All values are in microns; the numbers in brackets represent the 95% confidence intervals of the respective percentiles.
| Age 15 years | | |
|--------------|-------------------------|-------------------------|
| | 1st percentile (95% CI) | 5th percentile (95% CI) |
| Global | 76.0 (73.2 - 78.2) | 82.1 (79.9 - 83.9) |
| sector T | 46.9 (42.8 - 49.5) | 54.9 (51.7 - 57.0) |
| sector TS | 96.3 (90.8 - 100.0) | 107.4 (103.1 - 110.4) |
| sector TI | 99.1 (92.7 - 103.1) | 111.6 (106.6 - 115.0) |
| sector N | 38.3 (34.0 - 42.0) | 48.1 (44.6 - 51.0) |
| sector NS | 57.8 (52.0 - 62.6) | 70.7 (66.1 - 74.6) |
| sector NI | 53.6 (46.7 - 59.2) | 68.8 (63.4 - 73.3) |
Age 45 vears
### Age 65 years
| | 1st percentile (95% CI) | 5th percentile (95% CI) |
|-----------|-------------------------|-------------------------|
| Global | 74.5 (71.7 - 76.7) | 80.6 (78.4 - 82.4) |
| sector T | 43.9 (39.8 - 46.5) | 52.0 (48.8 - 54.1) |
| sector TS | 93.0 (87.6 - 96.8) | 104.1 (99.8 - 107.2) |
| sector TI | 94.6 (88.2 - 98.6) | 107.1 (102.1 - 110.4) |
| sector N | 38.3 (34.0 - 42.0) | 48.1 (44.6 - 51.0) |
| sector NS | 57.8 (52.0 - 62.6) | 70.7 (66.1 - 74.6) |
| sector NI | 51.9 (45.0 - 57.5) | 67.2 (61.7 - 71.7) |
3
{3}------------------------------------------------
# Agreement Study with Predicate
An agreement study was conducted to compare RNFL classification results from Spectralis HRA+OCT to results from Stratus in normal and glaucoma eyes. Healthy subjects (n=101) and glaucoma patients (n=183) were examined using Spectralis and Stratus. Results of this analysis indicate that there is a good linear correlation between RNFL thickness measurements with the Spectralis and Stratus devices for healthy and for glaucoma subjects and in all measurement regions. The slopes and intercepts of the regression lines are in the neighborhood of 1 and 0, respectively, although they vary in the different measurement regions. Because of these differences, Spectralis and Stratus RNFL thickness measurement results should not be used interchangeably. (This is in agreement with the published literature.)
#### Studies in Eyes with Disease
Case Reports and Case Series of eyes from patients with various pathologies were examined with Spectralis HRA+OCT and these images show that no artifacts were found and there were no unexpected RNFL thickness measurement results or unexpected classification results. RNFL thickness was found to be predictably decreased in glaucoma subjects.
{4}------------------------------------------------
Image /page/4/Picture/1 description: The image shows a logo for the U.S. Department of Health & Human Services. The logo consists of a stylized depiction of three human figures, arranged in a way that resembles a bird in flight. The figures are black against a white background. The text "DEPARTMENT OF HEALTH & HUMAN SERVICES - USA" is arranged in a circular fashion around the figures.
Food and Drug Administration 10903 New Hampshire Avenue Document Control Room -WO66-G609 Silver Spring, MD 20993-0002
Heidelberg Engineering GmbH c/o Diane Mandell-Horwitz, Ph.D., RAC Official Correspondent and Consultant The Weinberg Group, Inc. 1220 Nineteenth St., NW, Suite 300 Washington, D.C., 20036
OCT 01 2010
Re: K101223
> Trade/Device Name: Spectralis™ HRA + OCT Regulation Number: 21 CFR 886.1570 Regulation Name: Ophthalmoscope, AC-Powered Regulatory Class: II Product Code: OBO, MYC Dated: September 29, 2010 Received: September 29, 2010
Dear Dr. Mandell-Horwitz:
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 (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. 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.
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
{5}------------------------------------------------
Page 2 - Dr. Mandell-Horwitz
(21 CFR Part 807); labeling (21 CFR Part 801); medical device reporting (reporting of medical device-related adverse events) (21 CFR 803); good manufacturing practice requirements as set forth in the quality systems (QS) regulation (21 CFR Part 820); and if applicable, the electronic product radiation control provisions (Sections 531-542 of the Act); 21 CFR 1000-1050.
If you desire specific advice for your device on our labeling regulation (21 CFR Part 801), please go to http://www.fda.gov/AboutFDA/CentersOffices/CDRH/CDRHOffices/ucm115809.htm for the Center for Devices and Radiological Health's (CDRH's) Office of Compliance. Also, please note the regulation entitled, "Misbranding by reference to premarket notification" (21CFR Part 807.97). For questions regarding the reporting of adverse events under the MDR regulation (21 CFR Part 803), please go to
http://www.fda.gov/MedicalDevices/Safety/ReportaProblem/default.htm for the CDRH's Office of Surveillance and Biometrics/Division of Postmarket Surveillance.
You may obtain other general information on your responsibilities under the Act from the Division of Small Manufacturers, International and Consumer Assistance at its toll-free number (800) 638-2041 or (301) 796-7100 or at its Internet address http://www.fda.gov/MedicalDevices/Resourcesfor You/Industry/default.htm.
Sincerely yours,
Kesia Alexander
Malvina B. Eydelman, M.D. Director Division of Ophthalmic, Neurological and Ear, Nose and Throat Devices Office of Device Evaluation Center for Devices and Radiological Health
Enclosure
{6}------------------------------------------------
# Indications for Use
510(k) Number (if known): K101223
Device Name:
### Spectralis™ HRA+OCT
Indications For Use:
The Spectralis HRA+OCT is a non-contact ophthalmic diagnostic imaging device. It is intended for viewing the posterior segment of the eye, including two- and threedimensional imaging, cross-sectional imaging, fundus photography, and fluorescence imaging (fluorescein, indocyanine green and autofluorescence), and to perform measurements of ocular anatomy and ocular lesions. The device is indicated as an aid in the detection and management of various ocular diseases including: age-related macular degeneration, macular edema, diabetic retinopathy, retinal and choroidal vascular diseases, glaucoma, and for viewing geographic atrophy as well as changes in the eye that result from neurodegenerative diseases. The Spectralis HRA+OCT includes a retinal nerve fiber layer thickness normative database, which is used to quantitatively compare the retinal nerve fiber layer in the human retina to a database of Caucasian normal subjects; the classification result is valid only for Caucasian subjects.
Prescription Use X (Part 21 CFR 801 Subpart D) AND/OR
Over-The-Counter Use (21 CFR 801 Subpart C)
(PLEASE DO NOT WRITE BELOW THIS LINE-CONTINUE ON ANOTHER PAGE IF NEEDED)
Concurrence of CDRH, Office of Device Evaluation (ODE)
Daniel Kausen M.D.
(Division Sign-Off) Division of Ophthalmic, Neurological and Ear, Nose and Throat Devices
510(k) Number K101223
Two short videos show you everything — or skip straight to the written tutorial if you'd rather read. You can reopen this any time from the Tutorial button in the top bar.
Part 1 — Search, results, and everyday workflows 16 min
Part 2 — Embeddings: the galaxy map 3 min
1. Search: exact and fuzzy
Type a phrase like "coronary artery calcification" into the search box. You get two kinds of results. Exact results match the literal phrase — prefix searches work ("coronary artery calcificati") but suffix searches do not. Fuzzy results match on the meaning and intent of your phrase rather than the exact words, and are sorted by relevance score. Hover over the Exact or Fuzzy badge on any row to see exactly why it matched.
Use the checkboxes above the results to narrow: SaMD keeps only software-only devices, AI / ML keeps only devices with AI.
Exact vs. fuzzy search: what's the difference?
Exact matches on the literal phrase (prefix search works, suffix does not). Fuzzy matches on the meaning and intent of the phrase rather than the exact words. Hover over the badge on any row to see why it matched.
You search "coronary artery calcification" and want only software devices with AI. What two filters do you apply?
Narrow by SaMD (software-only devices), then narrow by AI/ML (devices with AI).
2. The results table
Scroll right in the results table. The intended use is extracted for you — no need to open the PDF. The device story gives a high-level snapshot of what the device does and how it's used. The AI Performance sub-table shows each output name, acceptance criteria, observed values, and development/test dataset descriptions — the same format Innolitics uses for regulatory strategy outputs, and the fastest high-level fingerprint of an AI device. It is AI-generated but has been very reliable in practice.
Where do you find a device's intended use without opening the PDF?
Scroll right in the search results table. The intended use column is extracted for you; no need to dig into the 510(k) summary PDF.
What does the AI Performance sub-table show, and why is it useful?
Output name, acceptance criteria, observed values, development dataset description, and test dataset description. It's the same format we use for regulatory strategy output and Fast 510(k) input, and the fastest high-level fingerprint of an AI device. AI-generated but reliable in practice.
3. Judging fuzzy relevance
Fuzzy results trail off in relevance as you scroll. Use three signals to decide how far down to go: the fuzzy badge explanations, the intended use column, and whether your target output (e.g., Cobb angle) still appears in the AI Performance sub-table. Once it stops appearing, you're past the relevant zone. A top hit with a low score (~0.4) and a stretched explanation is a hint the closest predicates are far away — the project may be headed for De Novo. Note the fuzzy search is a pattern match: it doesn't handle negation ("not") well, and hardware devices can appear — filter by SaMD/AI ML to cut them.
How do you judge how far down fuzzy search results to go?
Use the relevancy signals: the fuzzy badge explanations, the intended use column, and whether the target output (e.g., Cobb angle) still appears in the AI Performance sub-table. Once it stops appearing, results are trailing off in relevancy.
4. Device detail page: chat and citations
Click a device name to open its detail page: device facts on the left, a chat window on the right. Ask something like "Describe the training data". The answer carries little citation bubbles — click one to jump to the highlighted passage in the source PDF, so you can verify every AI answer against the document. There's also a Download PDF button for sharing.
How do you verify an AI chat answer on the device detail page?
Click the citation bubbles to jump to the relevant highlight in the source document.
Reading rule for every project: how many summaries do you read in full?
At least the three most relevant 510(k) or De Novo summaries, in full. After that, use targeted chat questions to confirm your memory quickly. The tool supports this professional habit — it doesn't replace it.
5. Side-by-side comparison
Select multiple rows in the results table (aim for under ~10), then open the PDF Viewer tab. Ask one question — it goes to all selected devices in parallel, each with citations. This is the fastest way to compare and contrast devices: training data, PCCP scope, how they handled adding new scanners, and so on.
What does the side-by-side PDF viewer mode do?
Select multiple devices, open the PDF viewer tab, and ask one question (e.g., "Describe the training data"). It queries all selected devices simultaneously with citations, so you can compare and contrast quickly.
6. Collections
With rows selected, go to the Collections tab and create a labeled collection (e.g., "Cobb Angle Project"). Reload that selection any time — before a client call, pull up the collection and ask questions across all of its devices at once.
How do you save a set of selected devices for later use?
Select the rows, go to the Collections tab, and create a labeled collection (e.g., "Cobb Angle Project"). You can reload the selection anytime and carry it into the PDF viewer and other tabs that support selections.
7. Product codes and the regulations tree
Click a product code in the results to jump to it in the regulations tree — identification text, sibling product codes, and devices you can open in a PDF viewer on the right. Click a regulation number to see its identification, special controls, and related product codes. You can also search by product code or regulation number at the top of the tree. Always read the special controls if any exist for your device — it broadens your search and sharpens pre-kickoff research.
What can you do from the regulations tree view?
Browse product codes and regulation numbers, read the identification text and special controls, browse sibling product codes, open device PDFs on the right, and search by product code or regulation number at the top of the tree.
8. Chart view
Click Show Chart and segment by regulation number (or product code) to see which regulations dominate your result set. Clicking a regulation takes you into the regulations tree. Great for spotting that most matches are, say, hardware laparoscopic devices — a cue to go back and filter.
How do you see which regulations dominate a search result set?
Click "Show Chart" and segment by Regulation Number. Clicking a regulation takes you to the regulations tree.
9. The predicate graph
Open the Predicates tab for a family-tree view of predicate relationships. Click a node to trace its parents and children; selections from search carry over pre-selected. Commonly predicated devices are worth reading — a lot of people predicated them for a reason. The visual lineage is also handy on client calls, e.g. to show how a predicate family evolved and justify why your predicate still holds.
In the predicate graph, why are commonly predicated devices worth reading?
A lot of people predicated them for a reason. Clicking a node traces parents and children, and selections from search carry over pre-selected.
10. Embeddings: the galaxy map
The Embeddings tab plots every matching document in a 2-D "galaxy map" where semantically similar devices cluster together. Hover or click clusters to explore, and let AI label the clusters for you. Embeddings beat product codes for grouping: two devices can carry different product codes (LLZ vs. QIH) yet do the same thing — the embedding captures the meaning of the intended use and device story. This is also exactly how retrieval-augmented generation (RAG) works under the hood, and it makes a great visual on client calls.
Try it yourself
Head to the search page and work through a few of these AI/ML fuzzy searches to build intuition: perivascular fat on CT · aortic valve calcification opportunistic screening on noncontrast CT · breast cancer prediction on digital pathology slides · autism detection · gestational age prediction · a hearing aid that can also detect a pulse · foundation model based analysis of ECG · large language models · penetration test. Watch how the relevance scores, intended use, and AI Performance tables tell you when results stop being meaningful.