Long distance image transfer: first results of its use in histopathological diagnosis

On Edge Computing for Remote Pathology Consultations and Computations

Telepathology aims to replace the pathology operations performed on-site, but current systems are limited by their prohibitive cost, or by the adopted underlying technologies. In this work, we contribute to overcoming these limitations by bringing the recent advances of edge computing to reduce latency and increase local computation abilities to the pathology ecosystem. In particular, this paper presents LiveMicro, a system whose benefit is twofold: on one hand, it enables edge computing driven digital pathology computations, such as data-driven image processing on a live capture of the microscope. On the other hand, our system allows remote pathologists to diagnosis in collaboration in a single virtual microscope session, facilitating continuous medical education and remote consultation, crucial for under-served and remote hospital or private practice. Our results show the benefits and the principles underpinning our solution, with particular emphasis on how the pathologists interact with our application. Additionally, we developed simple yet effective diagnosis-aided algorithms to demonstrate the practicality of our approach.

Diagnostic accuracy of second-opinion diagnoses based on still images

Second opinion of histological specimens is an important part of the daily routine in anatomic pathology practices. Today, extramural second opinion can be easily obtained by sending still images via an electronic network. The aim of this study was to examine the diagnostic accuracy of second opinion diagnosis based on still images selected from glass slides of 90 archived cases originally referred for extramural second opinion. Two pathologists together diagnosed first the still images (phase 1) and then the glass slides (phase 2). Phase 1 and phase 2 diagnoses were compared with the original second opinion diagnoses (OSODs). The pathologists achieved the same diagnostic results in phase 1 and in phase 2 measured against the OSOD, 67.8% (n = 61) and 68.9% (n = 62) complete agreement, respectively. In 29 cases in phase 1, the diagnoses were discordant with the OSOD. Three cases had incorrect benign diagnoses and 8 cases had incorrect malignant diagnoses. There were 8 false-negative diagnoses regarding malignancy, 6 false-positive diagnoses regarding malignancy, and 4 other discordant diagnoses. Eleven of the 29 discordant diagnoses could have had clinical implications. In interpreting these results, it is important to acknowledge the observer variability in diagnostic histopathology in general. In conclusion, the results support the concept of using still images to obtain second opinion diagnosis.

The image pyramid system--an unbiased, inexpensive and broadly accessible method of telepathology

Although computerised information technology, including the Internet is broadly used and globally accessible it is still not a significant form of professional communications in diagnostic histopathology. The high cost of interactive dynamic telepathology systems makes their use limited outside the richest economies. In contrast static telepathology systems are relatively cheap but in practice their information content can be heavily biased by the choice of images sent by the consulting pathologist. The degree of this bias may be regarded simply as the amount of information transferred to a remote location expressed as a percentage of the total information present in the histological sample. We refer to this as the percentage of explicit versus implicit information. Another major source of bias may be found in the information transmitted in written or verbal discussion with a remote consultant. We have developed a system of static telepathology, the image pyramid, which attempts to minimise bias by transferring all of the information in a section to the consultant. It is inexpensive and should prove to be widely accessible.

Telepathology overview: from concept to implementation

Telepathology is the practice of pathology at a distance by using video imaging and telecommunications. Significant progress has been made in telepathology. To date, 12 classes of telepathology systems have been engineered. Rapid and ultrarapid virtual slide processors may further expand the range of telepathology applications. Next-generation digital imaging light microscopes, such as miniaturized microscope arrays (MMA), may make virtual slide processing a routine laboratory tool. Diagnostic accuracy of telepathology is comparable with that of conventional light microscopy for most diagnoses. Current telepathology applications include intraoperative frozen sections services, routine surgical pathology services, second opinions, and subspecialty consultations. Three telepathology practice models are discussed: the subspecialty practice (SSP) model; the case triage practice (CTP) model; and the virtual group practice (VGP) model. Human factors influence performance with telepathology. Experience with 500 telepathology cases from multiple organs significantly reduces the video viewing time per case (P < .01). Many technology innovations can be represented as S-curves. After long incubation periods, technology use and/or efficiency may accelerate. Telepathology appears to be following an S-curve for a technical innovation.

Use of telepathology for routine surgical pathology review in a test bed in the Department of Veterans Affairs

BACKGROUND

Routine surgical pathology review by telepathology could be an important service component of multi-institutional pathology laboratory systems. Such service networks would increase access for rural hospitals without on-site pathologists to a broader range of pathology services on a daily basis.

METHODS

In this clinical trial, we analyzed the diagnostic accuracy, deferral rates, and viewing times of two generalist pathologists using a hybrid dynamic/store-and-forward (HDSF) telepathology (TP) system to render diagnoses in real time on 200 consecutive surgical cases. The objective was to assess the efficacy of TP in providing diagnostic surgical pathology services to a remote hospital without an onsite pathologist. Surgical pathology specimens underwent gross preparation by specially trained personnel. When appropriate, this was done under the video supervision of a telepathologist. For TP, glass slides were placed on the stage of a robotic microscope at the Iron Mountain (MI) Department of Veterans Affairs Medical Center (VAMC) (remote site); control of the motorized microscope was then transferred to a pathologist located 220 miles away at the Milwaukee VAMC (host site). For each case, the telepathologist had the option of rendering a diagnosis or deferring the case for later analysis by conventional light microscopy (LM). After the slides were read by TP and a surgical pathology report had been generated, the slides were transported to Milwaukee, where they were reexamined by the telepathologist using LM and then by the pathology group practice or, when there was no consensus, by an outside consultant to establish a "truth" diagnosis.

RESULTS

Compared with the consensus ("truth") diagnosis, clinically important and overall concordance were 99.0% and 97.4%, respectively, by TP, and clinically important and overall concordance were 100.0% and 98.5%, respectively, by LM. The deferral rate was 2.5%. Examining glass slides by HDSF telepathology took an average of 4.43 minutes per slide and 12.09 minutes per case.

CONCLUSION

The high diagnostic accuracy and low rate of case deferral support the proposal that an offsite pathologist using HDSF telepathology can substitute effectively for an onsite pathologist as a service provider.

Diagnostic accuracy of a rural live video telepathology system

Accuracy of diagnoses rendered using a live video telepathology network was assessed for permanent sections of surgical pathology specimens. To determine accuracy, telepathology diagnoses were compared with those obtained by directly viewing the glass slide using a standard microscope. A total of 294 cases were read via both telepathology and glass slide by attending pathologists at a tertiary care medical center. Overall accuracy was defined as exact concordance between diagnoses. Clinically insignificant differences in diagnoses were excluded to determine clinically significant accuracy. For the 285 cases with complete data, the overall accuracy for telepathology was 0.912 (95% confidence interval [CI], 0.872-0.941), whereas the overall accuracy for glass slide readings was 0.968 (95% CI, 0.939-0.985). This difference is statistically significant (p = 0.009). When focusing on clinically significant discrepancies, where the difference in diagnosis might affect therapeutic decisions, the video accuracy was only slightly less than the glass slide accuracy (0.965 [95% CI, 0.934-0.982] vs. 0.982 [95% CI, 0.957-0.994], respectively), but this difference is not statistically significant (p = 0.302). Most of the cases with clinically significant differences involved lesions with inherently high interobserver variation. Certainty of diagnosis did not differ between video and glass slide readings (p = 0.911), but there was an association between certainty of diagnosis and diagnostic accuracy for video (p = 0.003 for clinically significant accuracies). Based on these findings, we recommend when using this telepathology system that only preliminary diagnoses should be given in the following situations: for diagnostic areas with known high interobserver variability; when the consultant has any degree of uncertainty about the presence or absence of the lesion in question; and when there is insufficient experience using telepathology as a diagnostic medium.

Telepathology: frozen section diagnosis at a distance

Telepathology may be used to provide a frozen section service to hospitals without a department or institute of pathology. We have developed a telepathology system using the commercially available Integrated Services Digital Network (ISDN). The main software and hardware elements of our system are: Apple Macintosh workstations, a program for simultaneous transfer of image, voice and data, and a data bank for storage of patients' data and microscopic images. A picture instrument manager (PIM) makes remote control of microscopes or other instruments possible. The system connects the Department of Pathology of the University of Basel with the Regional Hospital of Samedan, 250 km away, and the Regional Hospital of Burgdorf, 100 km away. During a period of 20 months, frozen sections with the hospitals in Samedan and Burgdorf were performed in 53 patients. Between 54 and 58 s were required for the transfer of a diagnostic 8-bit grey level image containing 341 +/- 26.1 (standard error) kbytes (n = 13) or a diagnostic 24-bit colour image containing 165 +/- 16.9 kbytes (n = 40). Frozen section diagnosis was completed in 20-40 min. True-positive diagnoses of malignant tumours were achieved in 85.7% of cases (sensitivity = 0.857). No false-positive diagnosis was made. In 3 of the 53 cases telepathological diagnosis was not possible for technical reasons.

Current status of telepathology

Telepathology is moving from the experimental stage to become a regular feature of pathology practice. This has been made possible by technical advances in telecommunications and image processing. Since 1990 the University Hospital of Tromsø has provided local hospitals in northern Norway with a remote frozen section service and with access to video conferences for the review of microscopic findings and for the discussion of major diagnostic issues. Several other hospitals in Norway are now participating in this development and practical relations among pathology laboratories for the purpose of consultation and education will be the next step in the procedure. Similar developments in telepathology have taken place in other countries. Standardization of network and telepathology workstations will be needed before extensive international collaboration can be achieved. Progress in high quality video devices, high capacity telecommunication lines and improved image compression techniques will increase the usage of telepathology services and make them cost-effective. Thus, telepathology will contribute to the development of pathology services in the next century.

Telepathology with an integrated services digital network--a new tool for image transfer in surgical pathology: a preliminary report

We describe a low-cost telepathology system working via a commercial integrated services digital network (ISDN) and consisting of modular software and hardware elements. The main elements are Apple Macintosh workstations; a software program for the simultaneous transfer of pictures, voice, and data; and procedures for image processing and general administration of all the information generated. Additionally, the system allows remote control of any peripheral instruments by a "picture-instrument manager." The transfer rate is currently 64 kbit/s; it will be extended to 128 kbit/s (ISDN basic rate) in the near future and to 2 Mbit/s (ISDN primary rate) in the next 2 years. The system was tested by the regional hospital in Samedan, Switzerland, and the Department of Pathology, University of Basel, Basel, Switzerland, a distance of 250 km, by offering a remote frozen section service to the regional hospital in 16 cases. Fifty-four to 58 seconds were needed for the transfer of a diagnostic 8-bit grey-level image containing 341 (median value) +/- 26.1 (standard error) kbytes (n = 13) or a diagnostic 24-bit color image containing 165 (median value) +/- 16.9 (standard error) kbytes (n = 3). The time required for a diagnostic session was between 25 and 35 minutes.