STG data set: 91 sounds from 91 patients scheduled for echocardiography
3 (experienced cardiologists)
Proctor Harvey data set: 363 sounds from 232 patients
3 (experienced cardiologists)
Indications for Use
The Stethographics Heart STG System, is an electronic auscultatory device, intended to provide support to the physician in the evaluation of heart sounds in patients. The product will acquire and record the acoustic signals of the heart and analyze these signals. The analysis procedure will identify specific heart sounds that may be present. Identified sounds include S1, S2, and suspected systolic murmurs. The device is indicated for use in the clinical setting, by a physician or by trained personnel who are acting on the orders of a licensed physician. It in not intended as a sole means of diagnosis. The interpretations of heart sounds offered by the Heart STG System are only significant when used in conjunction with physician over-read as well as consideration of all relevant patient data.
Device Story
Heart STG System is software-based analysis tool for electronic stethoscopes; acquires/records acoustic heart signals (20-2000Hz) via compatible electronic stethoscope (e.g., Andromed iStethos, Cardionics e-Scope, 3M E-4000). Software runs on Windows PC; records 20-second .wav files. Algorithm processes signals to identify/mark S1, S2, and suspected systolic murmurs on waveform; calculates heart rate. Used in clinical settings by physicians/trained personnel. Output provides visual markers on waveform to aid physician auscultation; intended to support, not replace, clinical judgment. Benefits include standardized documentation and automated identification of potential murmurs to assist in cardiac assessment.
Clinical Evidence
Bench and clinical testing performed. Clinical evaluation used two datasets: 91 patient recordings (STG data set) and 363 sounds from the Proctor Harvey gold standard course. Performance compared against consensus of 3 cardiologists. STG data set: 83% overall agreement, 70% sensitivity for murmurs, 93% specificity. Proctor Harvey set: 82% overall agreement, 91% sensitivity, 62% specificity. Heart rate correlation with manual calculation was 0.96 and 0.91 for respective datasets. Results indicate performance comparable to inter-cardiologist agreement.
Technological Characteristics
Software-based analysis system for electronic stethoscopes. Input: 20-2000Hz acoustic signals via 16-bit mono .wav files. Sampling frequency: 8,000 Hz. Compatible with standard electronic stethoscopes. Operates on Windows-based PC. Connectivity: Standalone software application. No specific material standards or sterilization requirements listed as it is a software/accessory system.
Indications for Use
Indicated for use in clinical settings by physicians or trained personnel under physician orders to support evaluation of heart sounds in patients. Identifies S1, S2, and suspected systolic murmurs. Not for use as a sole diagnostic tool.
Regulatory Classification
Identification
Manual stethoscope: A mechanical device used to project the sounds associated with the heart, arteries, and veins and other internal organs. Electronic stethoscope: An electrically amplified device used to project the sounds associated with the heart, arteries, and veins and other internal organs.
Special Controls
*Classification.* Class II (special controls). The device, when it is a lung sound monitor, is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 870.9.
{0}------------------------------------------------
DEC 2 3 2005
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
K052283
115
### Stethographics, Heart STG System 510(k) Summary
#### Company:
Stethographics, Inc 1153 Centre Street Boston, MA
#### Owner:
Raymond Murphy 1153 Centre St. Boston, MA Phone: (617) 983-7259 Fax: 617-522-4156
#### Contact:
Rozanne Paciej Stethographics, Inc 1153 Centre Street Boston, MA
#### Date Prepared:
August 17, 2005
#### Name of Device:
Stethographics Heart STG System
#### Common Name of Device:
Electronic Stethoscope
#### Classification Name:
Heart STG System 21 CFR 870.1870
#### Predicate Devices:
The Stethographics Heart STG System is substantially equivalent to the Zargis' Acoustic Cardioscan (#K031517).
Stethographics 510(k)
{1}------------------------------------------------
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
#### Intended Use:
The Stethographics Heart STG is software intended to provide support to the physician in the evaluation of heart sounds in patients.
The product will acquire and record the acoustic signals of the heart and analyze these The product will acquire and reversity identify specific heart sounds that may be present. Identified sounds include S1, S2, and suspected systolic murmurs.
The device is indicated for use in the clinical setting, by a physician or by trained The device is indicated for use in the entired physician. It in not intended as a sole means of diagnosis.
The interpretations of heart sounds offered by the Heart STG System are only significant The microletonio of near with physician over-read as well as consideration of all relevant patient data.
#### Device Description:
The Heart STG is software intended to provide support to medical personnel in the The Freart STO is software mension vare records and analyzes an acoustic signal of the e valuation of near bounds - aids the healthcare provider by marking SI, S2, and suspected systolic murmurs on the waveform.
As a complete system the Heart STG system consists of an electronic stethoscope and As a complete system the rical by on a computer with Windows operating system. The user must provide a compatible electronic stethoscope and a computer.
#### Technological Characteristics:
The technological characteristics of the new device are compatible to those of the predicate devices.
#### Performance Data:
Both bench testing and clinical testing were performed to ensure that the device performs Doth ochen testing and chilied tooling in this pre-market notification includes descriptive as intended. The intended supplied in the echnological characteristics.
This information is provided pursuant to the requirements of the Safe Medical Devices Act of 1990 (SMDA).
{2}------------------------------------------------
| Device Specification | |
|---------------------------|--------------------------------------------------------------------------------------------------------|
| Recording frequency range | 20Hz to 2000Hz |
| Sampling frequency | 8,000 Hz |
| Data recording | Standard .wav (PCM) files at resolution of 16 Bit, Mono. |
| Recording time | 20 seconds |
| Acoustic sensor | Electronic stethoscope. Compatible models include Andromed<br>iStethos, Cardionics e-Scope, 3M E-4000. |
| Number of sensors | 1 |
## ----------------------------------------------------------------------------------------------------------------Device Snecification
# Heart Sound Analysis STG Performance Data
In order to evaluate the Heart STG with both common heart disorders and more rare disorders, two different sets of cardiac sounds were utilized. The first set of cardiac fare disorders, two allerentise of Heart STG to record 91 patients that were scheduled Sounds were obtained of assessment (referred to as the STG data set). The second set was for conocurdio graphy abbeart sounds provided in the Clinical Auscultation of the Cardiovascular System, by Proctor Harvey and David C. Canfield (referred to as the Cardiovasoular Systems of Proctor Harvey course is considered the gold standard for I roctor frai vey tata bething and covers a broad range of heart disorders (363 sounds total), both common and rare. All sounds were graded by three experienced cardiologists. total), both common and fare. In be ositive when 2 out of 3 cardiologists agreed on the presence of murmur.
The STG data set included 91 sounds from 91 patients that were scheduled for th echocardiography assessment. Average age was 61±17 years (range 23 to 94). The cender distribution in this population was 44% male and 56% female. The patients were gender distribution in this peparates recording. The STG recorded 20 second sounds asked to nota then orean ouring stres STG performance to that of the cardiologists. The '+' indicates murmur-positive sounds and '-' indicates murmur-negative sounds.
| | | Table 1. STG compared to 2 out of 3 cardiologists. STG Data Set. | |
|--|--|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--|
| | | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | |
| Number of sounds 91 | 2 out of 3 cardiologists + | 2 out of 3 cardiologists - |
|---------------------|----------------------------|----------------------------|
| Heart STG + | 23 | 4 |
| Heart STG - | 10 | 54 |
The corresponding overall percent agreement = 100%*(23 + 54)/91 = 83% The contesponding overall percent adjoonists -positive = 100%*(23)(23+10) = 70% Agreement of the STG with 2/3 cardiologists-negative = 100%*(54)/(54+4) = 93%
The Proctor Harvey data set included 363 sounds from 232 patients. Sixty-three sounds were recorded from children, 300 from adults. The Proctor Harvey data set is sounds were recorded from only of common and less common heart pathologies and, represemative of grout namber of be algorithm's performance in both common and less ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Stethographics 510(k)
400253 1 / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /
{3}------------------------------------------------
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------common heart conditions. Table 2 compares STG performance to that of the cardiologists.
## Table 2. STG compared to 2 out of 3 cardiologists. Proctor Harvey Data Set.
| Number of sounds 363 | 2 out of 3 cardiologists<br>+ | 2 out of 3 cardiologists<br>- |
|----------------------|-------------------------------|-------------------------------|
| Heart STG + | 227 | 43 |
| Heart STG - | 23 | 70 |
The corresponding overall percent agreement = 100%*(227 + 70)/363 = 82% The concesponaing & verair p 2/3 cardiologists-positive = 100%*(227)(227+23) = 91% Agreement of the STG with 2/3 cardiologists-negative = 100%*(70)/(70+43) = 62%
The agreement between cardiologists was also calculated, Tables 3 and 4, columns 3 to 8.
### Table 3. Agreement between cardiologists. STG Data Set.
| | STG<br>compared to<br>2 out of 3<br>cardiologists | Cardiologist<br>1 with<br>cardiologist<br>2 as a gold<br>standard | Cardiologist<br>1 with<br>cardiologist<br>3 as a gold<br>standard | Cardiologist<br>2 with<br>cardiologist<br>1 as a gold<br>standard | Cardiologist<br>2 with<br>cardiologist<br>3 as a gold<br>standard | Cardiologist<br>3 with<br>cardiologist<br>1 as a gold<br>standard | Cardiologist<br>3 with<br>cardiologist<br>2 as a gold<br>standard |
|------------------------------------------|---------------------------------------------------|-------------------------------------------------------------------|-------------------------------------------------------------------|-------------------------------------------------------------------|-------------------------------------------------------------------|-------------------------------------------------------------------|-------------------------------------------------------------------|
| Agreement<br>with<br>murmur-<br>positive | 70 | 91 | 100 | 72 | 93 | 67 | 79 |
| Agreement<br>with<br>murmur-<br>negative | 93 | 79 | 77 | 94 | 89 | 100 | 96 |
## Table 4. Agreement between cardiologists. Proctor Harvey Data Set.
| | STG<br>compared to<br>2 out of 3<br>cardiologists | Cardiologist<br>1 with<br>cardiologist<br>2 as a gold<br>standard | Cardiologist<br>1 with<br>cardiologist<br>3 as a gold<br>standard | Cardiologist<br>2 with<br>cardiologist<br>1 as a gold<br>standard | Cardiologist<br>2 with<br>cardiologist<br>3 as a gold<br>standard | Cardiologist<br>3 with<br>cardiologist<br>1 as a gold<br>standard | Cardiologist<br>3 with<br>cardiologist<br>2 as a gold<br>standard |
|------------------------------------------|---------------------------------------------------|-------------------------------------------------------------------|-------------------------------------------------------------------|-------------------------------------------------------------------|-------------------------------------------------------------------|-------------------------------------------------------------------|-------------------------------------------------------------------|
| Agreement<br>with<br>murmur-<br>positive | 91 | 97 | 97 | 83 | 94 | 82 | 93 |
| Agreement<br>with | 62 | 61 | 60 | 91 | 86 | 91 | 88 |
{4}------------------------------------------------
| | =469========================================================================================================================================================================== | | | |
|----------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--|--|--|
| | | | | |
| murmur- | | | | |
| negative | | | | |
Thus, the STG agreement with 2/3 cardiologists was found to be comparable to that measured among experienced cardiologists.
Heart Rate (HR) was validated by comparing the HR calculated by the STG to that Treat Rale (TIT) was vanuated by contpassion. The correlations between the Heart STG canculated by JN Entimall: 21.000 tlosscope for the STG data set and Proctor Harvey data set are 0.96 and 0.91, respectively.
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
41,30200 p 5 / 5
{5}------------------------------------------------
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
{6}------------------------------------------------
DEPARTMENT OF HEALTH & HUMAN SERVICES
Image /page/6/Picture/2 description: The image shows the logo for the U.S. Department of Health and Human Services. The logo features a stylized eagle head with three lines forming the head and neck. The words "DEPARTMENT OF HEALTH AND HUMAN SERVICES. USA" are arranged in a circular pattern around the eagle head.
DEC 2 3 2005
Food and Drug Administration 9200 Corporate Boulevard Rockville MD 20850
Stethographics, Inc. c/o Ms. Rozanne Paciej Director, Regulatory Affairs Quality Assurance 1153 Centre Street Boston, MA 02130
Re: K052283
Trade Name: Stethographics Heart STG System Regulation Number: 21 CFR 870.1875 Regulation Name: Stethoscope Regulatory Class: Class II (two) Product Code: DQD Dated: November 14, 2005 Received: November 22, 2005
Dear Ms. Paciej:
We have reviewed your Section 510(k) premarket notification of intent to market the device We have reviewed your Section 510(x) prematics is substantially equivalent (for the indications referenced above and have determined the actived predicate devices marketed in interstate for use stated in the encrosule) to regally manced promoted in vice Amendments, or to
commerce prior to May 28, 1976, the enactment date of the Federal Food. Drug commerce prior to May 28, 1970, the chacance with the provisions of the Federal Food, Drug, devices that have been reculted in accordance while approval application (PMA).
and Cosmetic Act (Act) that do not require approval of a providions of the Act. The and Cosment Act (Act) that do not require approvate of the general controls of the Act. The You may, therefore, market the device, subject to tits gonto for annual registration, listing of the and general controls provisions of the Free merals required.
devices, good manufacturing practice, labeling, and prohibitions against misbranding and adulteration.
If your device is classified (see above) into either class II (Special Controls) or class III (PMA), it If your device is classified (SCC above) the entile is a control stars and control device can
may be subject to such additional controls. Existing major regulations EDA mov may be subject to such additional controllar antitudes of the 21, Parts 800 to 898. In addition, FDA may be found in the Code of Poderal Regaring your device in the Federal Register.
{7}------------------------------------------------
Page 2 – Ms. Rozanne Paciej
Please be advised that FDA's issuance of a substantial equivalence determination does not mean Please be advised that FDA s issualice of a subscription with other requirements of the Act
that FDA has made a determination that your device complies . You must that FDA has made a decidination that your de received by other Federal agencies. You musti of any Federal statutes and regulations daminities, but not limited to: registration and listing (21)
comply with all the Act's requirements, including, but not limited to: r comply with all the Act s requirements, morams, connective requirements as set CFR Part 807); labeling (21 CFR Part 807); good if applicable, the electronic
forth in the quality systems (QS) regulation (21 CFR Part 820); and if applicable, the electroni form in the quality systems (QD) regulation (Sections 531-542 of the Act); 21 CFR 1000-1050.
product radiation control provisions (Sections 531-542 of the Act); 21 CFR 1000-1 product radiation control provisions (Decions of the as described in your Section 5 (0(k)
This letter will allow you to begin marketing your device as described in your high I his letter will anow you to begin manceing your and equivalence of your device to a legally premarket notification. The PDA multing of substantial equal and thus, permits your device to proceed to the market.
If you desire specific advice for your device on our labeling regulation (21 CFR Patt 801), please if you desire specific advice for your ac not one one one not the regulation entitled. comact the Office of Complanes at (210) = 15 = 16 = 16 = 1 = 1 = 1 = 1 = 1 = 1 = 5 = = \$ Small Misoraliung by reference to promantovitibilities under the Act from the Division of Small other general information on your responsibilities and its toll-free number (800) 638-204 or btrail Manufacturers, International and October http://www.fda.gov/cdrl/industry/support/index.html.
Sincerely yours,
A. Ui Feinter for
Bram D. Zuckerman, M.D.
Director Division of Cardiovascular Devices Office of Device Evaluation Center for Devices and Radiological Health
Enclosure
{8}------------------------------------------------
# Indications for Use
510(k) Number (if known): _ KOS 2283
K052283
Device Name: Stethographics Heart STG System
Indications for Use:
The Stethographics Heart STG System, is an electronic auscultatory device, intended to provide support to the physician in the evaluation of heart sounds in patients.
The product will acquire and record the acoustic signals of the heart and analyze these signals. The analysis procedure will identify specific heart sounds that may be present. Identified sounds include S1, S2, and suspected systolic murmurs.
The device is indicated for use in the clinical setting, by a physician or by trained personnel who are acting on the orders of a licensed physician. It in not intended as a sole means of diagnosis.
The interpretations of heart sounds offered by the Heart STG System are only significant when used in conjunction with physician over-read as well as consideration of all relevant patient data.
Prescription Use XXX (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 OF NEEDED)
Concurrence of CDRH, Office of Device Evaluation (ODE)
A.M. Mcclellan for Brian D. Zuckerman, (MD)
(Division Sign-Off)
Division of Cardiovascular Devices
510(k) Number
Page 1 of 1
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.