Computer vision and machine learning image analysis algorithms
—
—
—
—
—
—
Indications for Use
The Minuteful - kidney test is an in-vitro diagnostic, home-use urine analysis test system for the semi-quantitative measurement of albumin and creatinine in urine, as well as the presentation of their ratio, the albumin-creatinine ratio (ACR). The system consists of a smartphone application, proprietary Color-Board and an ACR Reagent Strip. The system is available for prescription-use only and is intended for people at risk of kidney disease. Results are intended to be used in conjunction with clinical evaluation as an aid in the assessment of kidney health.
Device Story
System uses smartphone camera to capture images of urine test strip and proprietary Color-Board; smartphone application processes images to perform semi-quantitative urinalysis for albumin and creatinine. Device operates in home setting; patient-operated. Application includes automated image quality checks; blocks image capture if lighting, blur, distance, angle, or strip placement are suboptimal. Output provides albumin-creatinine ratio (ACR) to aid clinicians in assessing kidney health. Modification from predicate involves enhanced control over camera exposure and white balance settings.
Clinical Evidence
No new clinical data provided. Substantial equivalence supported by analytical performance testing, including Limit of Detection (CLSI EP17-A2) and extensive Image Validation Transfer System (IVTS) studies evaluating performance under varied lighting, physical, and environmental conditions (shadows, blur, strip placement, and surface contamination). Previous clinical method comparison and analytical performance data from K210069 remain applicable.
Technological Characteristics
System uses reflectance photometry via smartphone camera and photosensitive diode. Components: ACR reagent strip (URiSCAN 2ACR), proprietary color-board, and smartphone app. Connectivity: Internet-enabled data transfer to backend server and EMR. Software: Computer vision and machine learning algorithms for image analysis and validation. Platform: Smartphone-agnostic via IVTS. Dimensions: 106mm x 143mm x 30mm; Weight: 0.105kg.
Indications for Use
Indicated for people at risk of kidney disease for home-use, prescription-only, semi-quantitative measurement of urine albumin and creatinine and calculation of albumin-creatinine ratio (ACR) to aid in assessment of kidney health.
Regulatory Classification
Identification
A urinary protein or albumin (nonquantitative) test system is a device intended to identify proteins or albumin in urine. Identification of urinary protein or albumin (nonquantitative) is used in the diagnosis and treatment of disease conditions such as renal or heart diseases or thyroid disorders, which are characterized by proteinuria or albuminuria.
{0}
FDA
U.S. FOOD & DRUG
ADMINISTRATION
# 510(k) SUBSTANTIAL EQUIVALENCE DETERMINATION DECISION SUMMARY
ASSAY AND INSTRUMENT
## I Background Information:
A 510(k) Number
K222921
B Applicant
Healthy.io Ltd.
C Proprietary and Established Names
Minuteful-kidney test
D Regulatory Information
| Product Code(s) | Classification | Regulation Section | Panel |
| --- | --- | --- | --- |
| JFY | Class II | 21 CFR 862.1225 - Creatinine Test System | CH - Clinical Chemistry |
| JIR | Class I | 21 CFR 862.1645 - Urinary protein or albumin (nonquantitative) test system | CH - Clinical Chemistry |
| KQO | Class I | 21 CFR 862.2900 - Automated urinalysis system | CH - Clinical Chemistry |
## II Submission/Device Overview:
A Purpose for Submission:
Modification to an existing device
B Measurand:
Albumin and Creatinine in urine
Food and Drug Administration
10903 New Hampshire Avenue
Silver Spring, MD 20993-0002
www.fda.gov
{1}
C Type of Test:
Semi-quantitative urinalysis
III Intended Use/Indications for Use:
A Intended Use(s):
See Indications for Use below.
B Indication(s) for Use:
The Minuteful - kidney test is an in-vitro diagnostic, home-use urine analysis test system for the semi-quantitative measurement of albumin and creatinine in urine, as well as the presentation of their ratio, the albumin-creatinine ratio (ACR). The system consists of a smartphone application, proprietary Color-Board and an ACR Reagent Strip. The system is available for prescription-use only and is intended for people at risk of kidney disease. Results are intended to be used in conjunction with clinical evaluation as an aid in the assessment of kidney health.
C Special Conditions for Use Statement(s):
Rx - For Prescription Use Only
D Special Instrument Requirements:
The following representative smartphone models and operating system (OS) versions were used in the different studies described in Section VII (Performance Characteristics).
| Manufacturer | Model | OS† version |
| --- | --- | --- |
| Apple | iPhone 8 plus | iOS16 |
| Apple | iPhone 11 | iOS13 |
| Apple | iPhone 13 Pro | iOS16 |
| Motorola | Moto G60S | Android 11 |
| Google | Pixel 5 | Android 13 |
| Samsung | Galaxy A42 | Android 12 |
| Samsung | Galaxy Note 10+ | Android 11 |
| Samsung | Galaxy 22 | Android 13 |
†Operating System
Phones and OS versions that are not compatible with the Minuteful – kidney test will be blocked from downloading the application.
K222921 - Page 2 of 8
{2}
K222921 - Page 3 of 8
## IV Device/System Characteristics:
### A Device Description:
The device description is the same as described in K210069. The device has been modified to increase control over the two camera features, exposure and white balance, during the image capturing process.
### B Principle of Operation:
The principle of operation is the same as described in K210069.
### C Instrument Description Information:
1. **Instrument Name:**
Minuteful - kidney test
2. **Specimen Identification:**
The specimen identification is the same as described in K210069.
3. **Specimen Sampling and Handling:**
The specimen sampling and handling is the same as described in K210069.
4. **Calibration:**
As described in K210069, calibration is not needed.
5. **Quality Control:**
Quality control is the same as described in K210069.
## V Substantial Equivalence Information:
### A Predicate Device Name(s):
Minuteful - kidney test
### B Predicate 510(k) Number(s):
K210069
### C Comparison with Predicate(s):
| Device & Predicate Device(s): | K222921 | K210069 |
| --- | --- | --- |
| Device Trade Name | Healthy.io Minuteful - kidney test | Healthy.io Minuteful - kidney test |
| General Device Characteristic Similarities | | |
| Intended Use/Indications For Use | The Minuteful – kidney test is an in-vitro | Same |
{3}
| Device & Predicate Device(s): | K222921 | K210069 |
| --- | --- | --- |
| | diagnostic home- use
urine analysis test
system, for the semi-quantitative
measurement of
albumin and creatinine
in urine, as well as the
presentation of their
ratio, the albumin-
creatinine ratio (ACR).
The system consists of a
smartphone application,
proprietary Color-Board
and an ACR Reagent
Strip. The system is
available for
prescription-use only.
Results are intended to
be used in conjunction
with clinical evaluation
as an aid in the
assessment of kidney
health. | |
| Detection Methodology | Reflectance photometry | Same |
| Detection Device | Photosensitive diode | Same |
| Measuring Range | 10-150mg/L albumin
10-300 mg/dL
creatinine | Same |
| General Device
Characteristic Differences | | |
| Camera Features | Advanced control of
white balance and
exposure | Without advanced
control |
VI Standards/Guidance Documents Referenced:
- CLSI EP 17-A2: Evaluation of Detection Capability for Clinical Laboratory Measurement Procedures, Approved Guideline - 2nd Ed.
- CLSI EP12-A2 2nd Edition, User Protocol for Evaluation of Qualitative Test Performance
K222921 - Page 4 of 8
{4}
VII Performance Characteristics (if/when applicable):
A Analytical Performance:
1. Precision/Reproducibility:
This was previously established in K210069.
2. Linearity:
This was previously established in K210069.
3. Analytical Specificity/Interference:
This was previously established in K210069.
4. Traceability, Stability, Expected Values (Controls, Calibrators, or Methods):
This was previously established in K210069.
5. Detection Limit/Cutoffs:
The sponsor collected a total of 57 urine samples from subjects with Diabetes Mellitus Type II, Diabetes Mellitus Type I, Hypertension, Dyslipidemia, Cardiovascular disease, a known kidney disease (e.g., Minimal Change Disease, Nephrolithiasis, IgA nephropathy), known Malignancy (e.g., Prostate, Lymphoma), Autoimmune disease, Tobacco use disorder, and obesity. The median age of this population was 61. The samples were pooled and diluted with negative urine to produce different levels of albumin and creatinine.
Albumin and creatinine were measured in test samples using a commercially available device that quantitatively measures albumin and creatinine in urine.
The sponsor conducted testing with 3 lots of their device. Several samples with albumin and creatinine levels around the cutoffs were tested and the sponsor used a Probit model described in the CLSI EP17-A2 guideline to estimate the C5, C50 and C95 for each of the cutoffs. Testing was conducted with 8 representative smartphones.
The sponsor’s study supports that the modification to the device did not change the cutoffs between the albumin and creatinine categories when the test is used with different smartphones and operating systems.
6. Assay Cut-Off:
Not Applicable.
7. Accuracy (Instrument):
This was previously established in K210069.
K222921 - Page 5 of 8
{5}
8. Carry-Over:
This was previously established in K210069.
B Comparison Studies:
1. Method Comparison with Predicate Device:
This was previously established in K210069.
2. Matrix Comparison:
Not Applicable. This test is only for urine samples.
C Clinical Studies:
1. Clinical Sensitivity:
Not Applicable.
2. Clinical Specificity:
Not Applicable.
3. Other Clinical Supportive Data (When 1. and 2. Are Not Applicable):
Not Applicable.
D Clinical Cut-Off:
Not Applicable.
E Expected Values/Reference Range:
This was previously established in K210069.
F Other Supportive Instrument Performance Characteristics Data:
The sponsor conducted several validation studies with urine samples spiked with albumin and creatinine to three levels of albumin, creatinine, and ACR. The expected results for the three levels of urine samples tested were 10 mg/L, 30-80 mg/L, and 150 mg/L for albumin; 10 mg/dL, 50 mg/dL, and 200-300 mg/dL for creatinine; Normal, Abnormal, and High-abnormal for ACR.
Illumination
Testing was conducted under different lighting conditions of different color "temperatures" and intensities from different light sources representative of those that may be used in the home setting. Testing was conducted with 8 representative smart phones of different models using
K222921 - Page 6 of 8
{6}
different operating systems. The results of the study support that the performance is not impacted by lighting conditions that are likely to be found in the home use environment.
## Boundary conditions- extreme illumination
a) Testing was conducted under different lighting conditions with different saturation of red, green, and blue light. Testing was conducted with 8 representative phones of different models and operating systems. Phones were tested under "boundary conditions" that were the maximum saturation of red, green, or blue light under which the device allowed the phone to take a picture of the urine test strip and color-board (under red, green, or blue light saturation that exceeded the boundary conditions, the application blocks the user from taking a picture). The results of this testing support that the device performance is not impacted by different light color saturation.
b) Testing was conducted under different light intensities (darkness and brightness measured in lux). Testing was conducted with 8 representative phones of different models and operating systems. Phones were tested under "boundary conditions" of light intensity that were the minimum light intensity and the maximum light intensity at which the device allows the phone to take a picture of the color board and urine test strip (under lighting conditions that exceeded the "boundary conditions" of light intensity i.e., darkness or brightness, the application blocks the user from taking a picture). The results of this testing support that the device is not impacted by different light intensities.
## Boundary conditions- Shadow
Testing was conducted with 8 representative smart phones of different models and operating systems under different shadow configurations of different intensity and covering different portions of the color-board and urine test strip. Six different shadow configurations and 2 different shadow intensities were tested. For 3 of the shadow configurations, the device blocked the user from taking a picture of the color-board and urine test strip. For the other 3 shadow configurations, the device performance was not impacted by the presence of shadows on the color-board and test strip.
## Boundary conditions- Blurry
Testing was conducted with 8 representative smart phones of different models and operating systems under 14 different conditions of blurred images consisting of different levels of focus and motion blur. Under conditions that the devices detected too much image blur, the user was blocked from taking a picture of the color-board and the urine test strip. Under conditions where the device allowed a picture of the color-board and urine test strip to be taken, the device performance was not impacted by image blur.
## Boundary conditions- Distance and Angle
Testing was conducted with 8 representative smart phones of different models and operating systems. Each phone was tested under "boundary conditions" that were 2 different distance conditions and 2 different angle conditions. The distance conditions were the longest distance from the color board and the shortest distance from the color board at which the device allowed the phone to take a picture of the strip and color board (at longer or shorter distances, the device
K222921 - Page 7 of 8
{7}
blocks the user from taking a picture). The angle conditions were the most acute and most obtuse angles at which the device allowed the phone to take a picture of the urine test strip and color board (at more acute or obtuse angles, the device blocks the user from taking a picture). The performance of the device was not impacted under these “boundary conditions”.
## Boundary conditions- Misplaced Urine Strip
Testing was conducted with 8 representative smart phones of different models and operating systems with 8 different urine test strip placements. Of the 8 different placements, 4 urine test strip placements were recognized as invalid by the device and the device blocked the user from taking a picture of the color-board and urine test strip. When the device determined the stick placement to be valid, variations in stick placement did not impact the performance of the device.
## Boundary conditions- Dirty Color-Board
Testing was conducted with 8 representative smart phones of different models and operating systems. Testing was conducted with 3 substances (urine, coffee, and ink) and 4 different “dirty” configurations covering different sections of the color-board and urine test strip. The device blocked the user from taking a picture of the color-board and urine test strip under “dirty” configurations that the device determined to be invalid. Under the “dirty” conditions the device determined to be valid, the performance of the device was not impacted.
The sponsor’s studies support that under these conditions, the modified device will correctly prevent unqualified images for downstream data analysis. For the qualified images, the performance of the modified device is equivalent to that of the predicate.
## VIII Proposed Labeling:
The labeling supports the finding of substantial equivalence for this device.
## IX Conclusion:
The submitted information in this premarket notification is complete and supports a substantial equivalence decision.
K222921 - Page 8 of 8
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.