Discovering hidden knowledge through auditing clinical diagnostic knowledge bases

“Electronic Pediatrician”, a non-machine learning prototype artificial intelligence software for pediatric computer-assisted pathophysiologic diagnosis ― general presentation

BACKGROUND

Knowledge-based systems (KBS) are software applications based on a knowledge database and an inference engine. Various experimental KBS for computer-assisted medical diagnosis and treatment were started to be used since 70s (VisualDx, GIDEON, DXPlain, CADUCEUS, Internist-I, Mycin etc.).

AIM

To present in detail the “Electronic Pediatrician (EPed)”, a medical non-machine learning artificial intelligence (nml-AI) KBS in its prototype version created by the corresponding author (with database written in Romanian) that offers a physiopathology-based differential and positive diagnosis and treatment of ill children.

METHODS

EPed specifically focuses on the physiopathological reasoning of pediatric clinical cases. EPed has currently reached its prototype version 2.0, being able to diagnose 302 physiopathological macro-links (briefly named “clusters”) and 269 pediatric diseases: Some examples of diagnosis and a previous testing of EPed on a group of 34 patients are also presented in this paper.

RESULTS

The prototype EPed can currently diagnose 269 pediatric infectious and non-infectious diseases (based on 302 clusters), including the most frequent respiratory/digestive/renal/central nervous system infections, but also many other non-infectious pediatric diseases like autoimmune, oncological, genetical diseases and even intoxications, plus some important surgical pathologies.

CONCLUSION

EPed is the first and only physiopathology-based nml-AI KBS focused on general pediatrics and is the first and only pediatric Romanian KBS addressed to medical professionals. Furthermore, EPed is the first and only nml-AI KBS that offers not only both a physiopathology-based differential and positive disease diagnosis, but also identifies possible physiopathological “clusters” that may explain the signs and symptoms of any child-patient and may help treating that patient physiopathologically (until a final diagnosis is found), thus encouraging and developing the physiopathological reasoning of any clinician.

Scalable and interpretable alternative to chart review for phenotype evaluation using standardized structured data from electronic health records

OBJECTIVES

Chart review as the current gold standard for phenotype evaluation cannot support observational research on electronic health records and claims data sources at scale. We aimed to evaluate the ability of structured data to support efficient and interpretable phenotype evaluation as an alternative to chart review.

MATERIALS AND METHODS

We developed Knowledge-Enhanced Electronic Profile Review (KEEPER) as a phenotype evaluation tool that extracts patient's structured data elements relevant to a phenotype and presents them in a standardized fashion following clinical reasoning principles. We evaluated its performance (interrater agreement, intermethod agreement, accuracy, and review time) compared to manual chart review for 4 conditions using randomized 2-period, 2-sequence crossover design.

RESULTS

Case ascertainment with KEEPER was twice as fast compared to manual chart review. 88.1% of the patients were classified concordantly using charts and KEEPER, but agreement varied depending on the condition. Missing data and differences in interpretation accounted for most of the discrepancies. Pairs of clinicians agreed in case ascertainment in 91.2% of the cases when using KEEPER compared to 76.3% when using charts. Patient classification aligned with the gold standard in 88.1% and 86.9% of the cases respectively.

CONCLUSION

Structured data can be used for efficient and interpretable phenotype evaluation if they are limited to relevant subset and organized according to the clinical reasoning principles. A system that implements these principles can achieve noninferior performance compared to chart review at a fraction of time.

The "Electronic Pediatrician (EPed®)" - A clinically tested prototype software for computer-assisted pathophysiologic diagnosis and treatment of ill children

BACKGROUND

Knowledge-based systems (KBS) are software applications based on a knowledge base and an inference engine. Various experimental KBS for computer-assisted medical diagnosis (CAMD) and treatment were started to be used since 70 s (VisualDx, GIDEON, DXPlain, CADUCEUS, Internist-I, Mycin etc).

OBJECTIVES

This article briefly presents and tests the "Electronic Pediatrician (EPed)", a prototype medical user-interface-based stand-alone KBS software created by the author (in Romanian language) that offers both a computer-assisted pathophysiologic diagnosis and treatment of ill children. EPed currently covers the most frequent (and mainly infectious) respiratory, digestive, renal and central nervous system pediatric diseases treated in the Department of Pediatric Infectious Diseases (DPID) of the County Emergency Hospital of Târgoviște town (Dâmbovița County, Romania).

METHODS

EPed uses a new type of pathophysiologically-based diagnosis algorithm, which combines the relative incidences of diseases and the degree of covering any clinical picture of any patient. EPed was tested by a retrospective study in which the author has chosen 34 clinical child cases diagnosed and treated in DPID in the period of pandemics. The diagnoses predicted by EPed for each tested patient in part was compared to the diagnoses established by the clinician for each patient, so that to test the diagnostic accuracy of EPed.

RESULTS

The real diagnoses of the 34 children tested were found to occupy an average list-index of 12 (varying from 1 to 57) in the lists of EPed-predicted diagnoses for each case in part (based on the clinical and paraclinical signs inputs), which is considered an acceptable result for a demo version.

CONCLUSIONS

The diagnostic accuracy of EPed was acceptable and is perfectible in the future by using other formulas of scoring which combine the relative incidences of diseases and the degree of covering the clinical picture of a patient. EPed is the first pathophysiologically-based KBS focused on general pediatrics (mainly on pediatric infectious diseases) written in Romanian.

Learning an expandable EMR-based medical knowledge network to enhance clinical diagnosis

Electronic medical records (EMRs) contain a wealth of knowledge that can be used to assist doctors in making clinical decisions like disease diagnosis. Constructing a medical knowledge network (MKN) to link medical concepts in EMRs is an effective way to manage this knowledge. The quality of the diagnostic result made by MKN-based clinical decision support system depends on the accuracy of medical knowledge and the completeness of the network. However, collecting knowledge is a long-lasting and cumulative process, which means it's hard to construct a complete MKN with limited data. This study was conducted with the objective of developing an expandable EMR-based MKN to enhance capabilities in making an initial clinical diagnosis. A network of symptom-indicate-disease knowledge in 992 Chinese EMRs (CEMRs) was manually constructed as Original-MKN, and an incremental expansion framework was applied to it to obtain an expandable MKN based on new CEMRs. The framework was composed by: (1) integrating external knowledge extracted from the medical information websites and (2) mining potential knowledge with new EMRs. The framework also adopts a diagnosis-driven learning method to estimate the effectiveness of each knowledge in clinical practice. Experimental results indicate that our expanded MKN achieves a precision of 0.837 for a recall of 0.719 in clinical diagnosis, which outperforms Original-MKN and four classical machine learning methods. Furthermore, both external medical knowledge and potential medical knowledge benefit MKN expansion and disease diagnosis. The proposed incremental expansion framework sustains the MKN learning new knowledge.