Table 2 Publications in pathology CDSS implementation

CMDX©-based single source information system for simplified quality management and clinical research in prostate cancer [ 38 ]

Prostate Cancer

Created a system used to store topographical information about prostate sample biopsy specimens to create heat-maps for areas of most likely to be prostatic carcinoma, essential for clinical decision-making, prognosis, and research.

Between 2010 and 2011, generated 259 biopsy case reports uploaded to the database with 100% data completeness and a source-to-database error of 10.3 per 10,000 fields. Serves as an implementation of pathology data sharing through a healthcare information system.

A Teaching Database for Diagnosis of Hematologic Neoplasms Using Immunophenotyping by Flow Cytometry [ 83,84 ]

Hematologic Neoplasms

Developed of web-enabled relational database, by pooling literature for cell surface marker definitions of 37 hematologic neoplasms. Using this expression profile, an algorithm was created by pathologists to assist in teaching flow cytometry diagnosis of hematologic neoplasms

Algorithm for identifying hematologic marker expression patterns validated using 92 clinical cases with an identification success rate of 89%. Tool has been used by pathologists-in-training to develop flow cytometry interpretation skills.

Bayesian belief network for the Gleason patterns in prostatic adenocarcinoma: development of a diagnostic decision support system for educational purposes [ 85 ]

Prostate Cancer

Developed a Bayesian belief network (BBN) for Gleason grading of prostate adenocarcinoma, to allow subjective evaluation of prostatic carcinoma slides by computer.

As histological diagnosis of prostate carcinoma is often produces wide inter-interpreter variability. This tool serves as a decision support tool to interpret descriptive terms in pathology reports and accurately determine tumor grading.

A computer-based diagnostic and prognostic system forassessing urinary bladder tumour grade and predicting cancer recurrence [ 86 ]

Bladder Cancer

Designed a decision support system, employed by pathologists in microscopic observation of tissue samples and measurements of nuclear characteristics, allowing automatic assessment of urinary bladder tumor grade and cancer recurrence probability.

The system employed classified tumors with an accuracy of 82%, 80.5%, and 93.1% for tumors of grade I, II, and III. Suggested prognosis in 72.8% of samples with a confidence of 74.5%.

Comprehensive graphic-based display of clinical pathology laboratory data [ 87 ]

Information Sharing

Established a graphic-based computerized system for display of clinical pathology data to permit improved data access and sharing.

By displaying laboratory data in a graphical manner, designers aimed to create a user-friendly system meant to highlight pertinent trends necessary for decision management.

Electronic reminders for pathologists promote recognition of patients at risk for Lynch syndrome: cluster-randomised controlled trial. [ 88 ]

Colorectal cancer

Deployed a new guideline to improve recognition of patients at risk for Lynch syndrome by a multidisciplinary team of surgeons, pathologists and clinical geneticists.

An electronic reminder system for pathologists increasing identification of patients at high risk for Lynch syndrome by 18% when compared to the control arm.

Image-guided decision support system for pathology [ 89 ]

Hematologic Neoplasms

Developed an image based retrieval system containing 261 digitized images of lymphocyte disorders and regular lymphocytes. Queries to the system extract features of the histopathological images for comparison and identification.

Based solely on morphological characteristics, the analysis system performed better than humans in identifying certain lymphoproliferative disorders according to a ten-fold cross-validated confusion matrix.