MétaCan
Menu
Back to cohort
Record W2148580193 · doi:10.1186/1746-1596-8-s1-s10

Telepathology consultation in China using whole slide image and an internet based platform

2013· article· en· W2148580193 on OpenAlex

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

affAt least one author lists a Canadian institution in the pinned OpenAlex snapshot.

Bibliographic record

VenueDiagnostic Pathology · 2013
Typearticle
Languageen
FieldComputer Science
TopicAI in cancer detection
Canadian institutionsUniversity of British Columbia
Fundersnot available
KeywordsTelepathologyThe InternetPathologyMedicineComputer scienceChinaMedical physicsWorld Wide WebTelemedicineGeography

Abstract

fetched live from OpenAlex

Telepathology is a particularly useful pathological tool suitable for developing countries such as China, which generates a large number of pathology specimens each year due to the size of its population but has a shortage of well trained and experienced pathologists in primary cares hospitals and hospitals within under-developed regions. Pathologists in these hospitals often have difficulty diagnosing challenging pathology cases. Telepathology, especially second opinion or teleconsultation is one of the solutions to this problem [ 1 – 3 ]. We reported an internet based open telepathology consultation platform using whole slide images (WSI) in China. The results from the telepathology consultation service since inception have been analyzed and summarized for this publication. We believe that our experiences will help to promote telepathology in China and in other developing countries. The telepathology consultation cases used in this report represent pathology consultation sent to the telepathology consultation platform ( http://www.mpathology.cn/mpcc/ ) from the beginning of the service in July 2008 to May 30, 2011. The cases submitted for teleconsultation were from 29 institutions, which were equipped with a virtual microscope, Motic Virtual Microscopic Scanner (Motic, China). The equipment and related software used in this report were validated in a previous telepathology study using a variety of 600 surgical pathology specimens [ 4 ]. When the participating hospital had a pathology case requiring consultation, a referring pathologist logged into the website, http://www.mpathology.cn/mpcc/ with a secure user name and password and filled out an online request form which included the patient’s name, age, and relevant clinical information, gross findings, immunohistochemistry results, a preliminary diagnosis and the name of expert pathologist chosen for consultation. Referring pathologist then scanned, uploaded and attached WSI of one or several representative H/E slides as well as relevant immunohistochemistry stained slides to the request form, and then sent these to the platform. An internet based telepathology platform was used and a server was used for storage of images and data. The system was maintained by two information technology (IT) technicians and one system manager. When referring pathologists from submitting hospital sent request for consults to the platform, IT technicians would be alerted by e-mails. The technician would then exam the completeness of submitted materials and system manager would contact the expert pathologist for consultation. A panel of 84 Chinese pathology experts was invited and agreed to participate in the teleconsultation service. The names, affiliation of the pathologists and the areas of their subspecialty expertise were listed on the website ( http://www.mpathology.cn/mpcc/ ). The platform covered the medical malpractice insurance for the expert consultants. When an expert pathologist was requested for consultation, the expert was instantly noticed by the cell phone message and e-mail. The expert pathologist would use a computer or an Ipad to log into the website, review WSI, capture a representative image, write the pathologic diagnosis, sign the report with an electronic signature and then release the final report. Once the final report was released, the system sent an email to alert the system manager who then sent the final consultation report by e-mail or fax to the referring pathologist. Statistical analysis was performed using SPSS 7.0 edition. The number of cases submitted increased from 17 cases in 2008 to 587 cases in 2010. 935/1022 cases (91%) were sent in 2010 and the first 5 months of 2011. A total of 29 hospitals were participated in the telepathology consultation service. The number of participating hospitals was 3 in 2008, 9 in 2009, and 21 in 2010; a 7-folds increase from 2008 to 2010. The mean number of cases submitted by each hospital was 35 cases in the past 3 years. Ten hospitals accounted for 90% of the cases submitted, with the other 19 hospitals accounting for only 10% of the submitted cases. The average number of WSI per case was 1.8 with a range from 1 to 11. 62.0% of the cases only had one WSI, 27.6% had two WSI, and 10.4% had 3 or more WSI. 203 (19.8%) of the cases had included immunohistochemistry slides. The average time needed from transmitting teleconsultation request form with attached WSI image to the server to the release of teleconsultation pathology report by expert consultants was 38 hours with a range of less than 1 hour to 323 hours. The teleconsultation report was released within 12 hours for 43.2% of cases; 24 hours for 65.9% of cases and 48 hours for 79% of cases (Table 1 ). Although a list of 84 expert consultants was available for teleconsultation, only 43 expert consultants were requested by submitting hospitals for teleconsultation. Among these 43 experts, 23 consultants were requested most frequently, accounting for 95% of the consultation cases. The common sites of the pathology were gynecologic (24.2%), gastrointestinal/ liver/pancreatic (14.7%), and lung (13.1%), each accounting for more than 10% of cases. 810 (79.3%) out of 1022 were neoplastic pathology. Among them, 193 (21.3%) were benign tumors but 637 (78.7%) were malignant. 302 (29.5%) out of 1022 cases were not given a preliminary pathology diagnosis by submitting hospitals. Among 720 cases with a preliminary pathology diagnosis, 122 (16.9%) of the cases received a consultation report which was not in agreement with the preliminary pathology diagnosis (Table 2 ). Local hospital pathologists could not render a preliminary diagnosis or made a wrong preliminary diagnosis in 424 cases, representing 41.5% of the total teleconsultation cases.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.000
metaresearch head score (Gemma)0.001
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.504
Threshold uncertainty score0.615

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0000.001
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.000
Science and technology studies0.0000.000
Scholarly communication0.0000.001
Open science0.0000.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.015
GPT teacher head0.268
Teacher spread0.253 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it