The Reynolds Pathfinder 700 Holter Analyzer is intended to be used to analyze magnetic tapes of ambulatory electrocardiograms made on compatible Holter Recorders. The system will detect various arrhythmias and will measure ST elevation or depression. The system acts in both an automatic mode (tape analyzed without operator intervention) and in an interactive mode (operator can intervene and affect analysis).
Device Story
Pathfinder 700 is a high-speed Holter ECG analyzer; processes magnetic tapes or solid-state recorder data. System utilizes parallel computing architecture with 27 transputers (T400/T425) to achieve 1000x real-time analysis speed; 24-hour tape analyzed in ~1.5 minutes. Operates via tower cabinet with CRT display, keyboard, and mouse; GUI-based interface. Features include arrhythmia detection, ST deviation analysis, and pacemaker beat detection. Used in clinical settings by trained operators. Workflow: data extraction, EMG filtering, ECG database storage, filter chain smoothing, trigger positioning, and morphology classification. Output displayed on screen for automatic or interactive review; operator can accept or modify classifications. Benefits include rapid, automated cardiac rhythm assessment and efficient clinical review of ambulatory ECG data.
Clinical Evidence
Clinical testing compared Pathfinder 700 analysis against Pathfinder 3 and Marquette 8000 using standardized AHA test tapes. Non-clinical testing included extensive software validation and compliance with IEC 950 (1988) and EN 60590 (1986) safety standards, and EN 55022 (Level B) electromagnetic compatibility testing.
Technological Characteristics
Tower-based workstation; 486 motherboard for I/O; 27-transputer (T400/T425) parallel processing network; 188MB RAM; 1000MB hard disk. Inputs: magnetic tape/solid-state Holter recordings. Connectivity: mouse, keyboard, CRT display, laser printer. Software: DOS-based, icon-driven GUI. Analysis: Neilson algorithm-based. Standards: IEC 950, EN 60590, EN 55022.
Indications for Use
Indicated for analysis of ambulatory electrocardiogram (Holter) recordings to detect arrhythmias and measure ST segment changes in patients undergoing cardiac monitoring.
Regulatory Classification
Identification
A programmable diagnostic computer is a device that can be programmed to compute various physiologic or blood flow parameters based on the output from one or more electrodes, transducers, or measuring devices; this device includes any associated commercially supplied programs.
{0}
Reynolds Pathfinder 700 510(k) Summary Page B1
Reynolds Medical Ltd.
510(k) Submission
Pathfinder 700 Holter Analyzer
510(k) Summary
K951902
JUL 15 1995
(a) Basic Data
(1) Submitter Information
Name: Reynolds Medical Ltd.
Address: 1-2 Harforde Court, John Tate Road,
Hertford, Herts, SG13 7NW, England
Contact Person: Dr. George Myers, 201-438-2310
Date Prepared: April 1, 1995
(2) Names of Device
Proprietary Name: Pathfinder 700 Holter Analyzer
Common/Usual Name: Holter analyzer
Classification Name: Arrhythmia Detector and Alarm
(3) Predicate Devices:
The principal predicate device is the Reynolds Pathfinder 3, which received marketing permission from the FDA under K871344 on October 20, 1987. The principal modification of the Pathfinder 700 is the use of modern, high-speed, digital technology.
Other predicate devices are the DelMar 263 (a PC-based analyzer, and the Zymed 1510 (also a stand-alone analyzer). Comparisons of the Pathfinder to these devices are in the Comparison section.
(4) Description of the Device
I. Introduction
The Pathfinder 700 Holter ECG analyzer is a high-speed Holter cassette analyzer capable of analyzing tapes recorded at 1.47 mm/second and 1.0 mm/second as well as solid-state recorders such as the Reynolds eRAM and the Braemar DL700. It is a successor to the Reynolds Pathfinder 3 analyzer, marketing of which has been approved by the FDA under 510(k) number K871344. The unit features arrhythmia analysis, ST deviation analysis, pacemaker beat detection, and determination of beat types, in both an automatic and interactive mode. The system uses an icon-based, mouse-controlled, Graphical User Interface ("GUI"), visually
{1}
Reynolds Pathfinder 700 510(k) Summary Page B2
similar to those found on the well-known Apple and Windows operating systems. Analysis is performed at 1000 times the recording speed, obtained by using parallel computations and a 188 megabyte random-access memory, in addition to a 1000 megabyte hard disk. Thus, a 24 hour Holter tape can be analyzed in approximately 1.5 minutes. A 486 motherboard is used to provide disk control and keyboard/mouse interfacing, while the analysis sections use parallel computing, based on the T400 and T425 transputers, a computer element which forms the basis of many parallel computing systems. The Pathfinder system is mouse-controlled, and uses a cathode-ray tube display and keyboard. The analysis is based on the Neilson principle or algorithm, which have been used in all previous Reynolds analyzers, and which have received 510(k) approval. The basic principles of the analysis are the same as those used previously by Reynolds; the new features of this system lie principally in its new electronic features, which have led to re-programming the equations and analysis principles.
## II. Description of System
Physically, the analyzer computer is in a "tower" cabinet, to which are connected the keyboard, the mouse, the cathode-ray tube display, a laser printer. The tape reader is an integral part of the cabinet.
An optical disk accessory for archival storage is also offered. This option, which must be installed by Reynolds personnel, plays no role in the analysis or the operation of the program as is the solid state recorder interface.
The heart of the system, responsible for all Holter analysis, is the transputer data processing components. The transputer network communicates to peripherals by means of a 486 motherboard which gives it access to the hard disk and the keyboard. The tape reader, 188 megabyte RAM memory, video display, and laser printer communicate directly with the transputer components. The 486 motherboard boots the transputer network (including loading the individual RAMs with program) and then acts as a slave input/output processor for the keyboard, disk, mouse, and network services. It also monitors the "error" line from the transputer hardware and resets it if an error occurs. Network options are also available for archival storage.
The basic element of the parallel computer system is the transputer (derived from TRANSistor and comPUTER). The Pathfinder 700 uses mainly the T400 and T425 transputers (27 in all). Each of these devices have the basic capacity of
{2}
MEDSYS, INC.
201 438 3213
P.04
Reynolds Pathfinder 700 510(k) Summary Page B3
two INTEL 80386 chips (20 MHz), and each has internal memory. Basically, the transputer is an independent computing element (with the power of a motherboard chip with memory) which has links to other transputers, each of which has its own memory for computation and programming. Thus, they can each carry on independent computations. In addition, the transputers have internal timing circuits.
While this overview cannot give a detailed description of how parallel computing works, the following presents the general idea for a Holter monitor. The tape or solid state reader transfers the electrocardiogram to RAM, and the data in RAM is immediately analyzed by the parallel system while the tape-reading process is going on. The criteria for the analysis are set by the user, as will be explained later. The ECG data is also displayed and (if desired) printed and stored on disk. The results of the analysis are also stored in results storage (RAM), displayed, and printed and/or stored on disk. The mode of display can be selected by the user. However, since the entire operation of (tape reading > analysis > storage) takes only 90 seconds, the user is in general unaware of this underlying process. If the user has selected the "Automatic" mode, the results display and print-out can proceed after the 90 seconds. If "Interactive" mode is desired, the screens start showing intermediate results very rapidly, but the user is basically unaware of the fact that the entire analysis has probably already been completed.
## III. General Program Flow
This section will present a brief description of the general program flow as seen by the operator. Note that this section does not describe the actual internal workings of the analyzer.
When the power is turned on after a boot period, the operator first sees a "basic display screen."
The basic display screen has a menu bar at the top, a representation of the ECG section currently being analyzed in the center, and an expanded section of the ECG at the bottom. This expanded ECG is in a rectangle called the "Highlight Box," and is used for selecting segments of the ECG during interactive analysis. The analysis criteria, scale factors, etc., can be changed by the operator, as explained in the instruction manual. The section marked "mouse buttons" shows the functions of the buttons on the mouse, which may change during various parts of the analysis.
{3}
TESYS, INC.
201 438 3213
P.02
Reynolds Pathfinder 700 510(k) Summary Page B4
The first menu to be selected would be "Start Analysis," which resets all adjustable parameters to "power-up" status and clears the memories, including patient details. The parameters can then be changed by means of "Control Analysis," which sets various values to be used in the analysis: for example, sensitivity, criteria for various tests, the number of channels on the tape, and whether automatic or interactive analysis is desired. This can only be set through "START ANALYSIS." Since parameters are rarely changed for different patients, the values of these parameters may be stored so that they become the "power-up" parameters. However, parameters can be changed after starting the tape and the initial analysis with only a slight loss in time, since the procedure is so rapid. In interactive analysis, the operator is presented with each event or complex (as selected by the operator) in the sequence indicated, accompanied by the diagnosis performed by the analysis. The operator then has the choice of accepting or changing these diagnoses. The operator can always enter interactive analysis at any time. If the criteria are not to be changed for succeeding tapes, the set-up can be saved to disk and it will be unnecessary to enter this menu for succeeding tapes.
The "Display Menu" controls the nature of the various displays as the analysis proceeds. For an automatic analysis, these displays are unnecessary, but for an interactive analysis, the operator can regulate the displays to facilitate the interactive process.
When "Start Analysis" is selected, menus appear to set up the particular analysis to be performed, including the type of recorder, the analysis type (automatic or interactive), options for printing the report, and the start time of the analysis. "Analyze" starts the analysis itself.
The analysis performed is the same for "automatic" and "interactive" modes, and, as noted above, the analysis is completed in about 90 seconds. In automatic mode, the report prints the first analysis "as is." In interactive mode, the operator can either change the classifications of arrhythmias proposed by the analysis algorithm, or can define new morphologies to be identified by the system.
The "Recorder Type" menu lists a large number of the most common recorders, as well as "generic" recorders. The system will automatically adjust to the recording speed of the recorder selected (1.0 mm/sec., 1.47 mm./sec or others) from the menu, as well as special formats which may be used.
{4}
MENYS, INC.
201 438 3213
P.03
Reynolds Pathfinder 700 510(k) Summary Page B5
After the analysis, the "Report" menu can be used to control whether the report is printed or stored, its language (English or other), its format, the sequence of the events in the report, and the sections to be included. Again, if all tapes are to have the same report choices, this section does not have to be changed for each analysis.
The overall flow is the following: After the initial selection of parameters and options, the system enters either automatic or interactive analysis. These analyses have an arrhythmia section, and an ST section. The ST section is optional, but the arrhythmia detection must always be selected. The unit then continues with the analyses, which are explained in more detail in the software section.
In operation, the tape reader supplies data to the Data Extraction unit, which also provides EMG (muscle noise) filtering. The data is then put into an ECG database. There are two identical channels for each channel of recorded data, plus a third channel to read the pacemaker channel. The data first goes through a "filter chain" to smooth the data, and then goes to a unit which determines the "trigger" position (where the unit "triggers" on a complex and a shape classification processor. The outputs are then combined and the arrhythmias are classified. The results go to the graphical interface for display. The Video Display Control is the path by which the user makes modifications in interactive analysis. The 486 motherboard is under the control of the DOS operating system. There is no "operating system" as such for the transputers.
(5) Intended Use
The Reynolds Pathfinder 700 Holter Analyzer is intended to be used to analyze magnetic tapes of ambulatory electrocardiograms made on compatible Holter Recorders. The system will detect various arrhythmias and will measure ST elevation or depression. The system acts in both an automatic mode (tape analyzed without operator intervention) and in an interactive mode (operator can intervene and affect analysis).
(6) Comparison with Predicate Devices
The Pathfinder 700 is substantially equivalent to the Reynolds Pathfinder 3, the DelMar 263, and the Zymed 1510, as noted previously.
{5}
Reynolds Pathfinder 700 510(k) Summary Page B6
## (b) Performance Data
The Pathfinder 700 has been tested in comparison with the predicate devices, and also with standardized MIT and AHA tapes.
## (1) Non-clinical tests
The Pathfinder 700 has been tested by independent laboratories, and it meets the requirements of IEC 950 1988 with amendments 1 and 2 and EN60590 1986 with amendments 1 and 2. Since the device is not patient connected, it is classed as office equipment. Electromagnetic compatibility tests have been done to EN 55022 level B.
These analyzers are now manufactured for sale in the United Kingdom and sold in more than ten countries around the world. The system has also received a French homologation.
The software has undergone extensive validation testing. The tests and results are in the Software section.
## (2) Clinical Tests
The system has undergone a clinical test in which the analysis of the Pathfinder 700 has been compared to the analysis provided by a predicate device, the Pathfinder 3, and the Marquette 8000. In this test, the units under test analyzed standard AHA test tapes.
## (3) Conclusion
The conclusions drawn from the non-clinical and clinical tests demonstrate that the device is as safe and effective, and performs as well or better than the legally marketed devices identified in paragraph (a)(3).
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.