Journal: Health Data Science

Article Title: Cognitive Computing-Based CDSS in Medical Practice

doi: 10.34133/2021/9819851

Figure Lengend Snippet: A general framework of cognitive computing-based CDSS. Medical big data manages the pool of heterogeneous medical data and provides the universal data access interface towards upper layers. Perception provides capabilities to process medical data of various modalities as well as holistic multimodal data analysis. Cognition computerizes medical knowledge with explicit and implicit methods, which enables machine intelligence to process medical data like humans. Applications provide various AI capabilities for clinical decision support based on the results from former layers. Examples include diagnostic suggestion, detection of misdiagnosis, and treatment recommendation.

Article Snippet: Among the providers of cognitive computing-based CDSS, Baidu, iFLYTEK, and Dr. Mayson are the representatives of the real systems in China.

Techniques: Diagnostic Assay

Journal: Health Data Science

Article Title: Cognitive Computing-Based CDSS in Medical Practice

doi: 10.34133/2021/9819851

Figure Lengend Snippet: An example of diagnosis inference under the general framework of cognitive computing-based CDSS. Multimodal data about the patient are perceived by information systems. After extracting features and transforming to machine-readable medical knowledge, the features and knowledge acquired from different data sources are jointly analyzed to infer diagnosis.

Article Snippet: Among the providers of cognitive computing-based CDSS, Baidu, iFLYTEK, and Dr. Mayson are the representatives of the real systems in China.

Techniques: Biomarker Discovery

Journal: Health Data Science

Article Title: Cognitive Computing-Based CDSS in Medical Practice

doi: 10.34133/2021/9819851

Figure Lengend Snippet: The overview of the general clinical pathway as well as the intervention points of cognitive computing-based CDSS. Throughout the clinical path, cognitive computing-based CDSS provides various functions for clinical decision-making. For example, consultation support guides physicians when inquiring with patients by suggesting questions to physicians. Voice medical record frees physicians’ hands by automatically recording and translating physicians’ voice data into the texts in EMR documents. Diagnostic suggestion shows the likelihood of the most probable diseases that the patient suffers according to clinical notes.

Article Snippet: Among the providers of cognitive computing-based CDSS, Baidu, iFLYTEK, and Dr. Mayson are the representatives of the real systems in China.

Techniques: Diagnostic Assay

Journal: Health Data Science

Article Title: Cognitive Computing-Based CDSS in Medical Practice

doi: 10.34133/2021/9819851

Figure Lengend Snippet: The demonstration of cognitive computing-based CDSS developed by Baidu. When a physician submits the clinical document in the EMR system, CDSS perceives the data and provides clinical decision support according to the stage that the physician is in. When illness description (e.g., chief complaint and history of present illness) is entered but the diagnosis is empty, CDSS provides diagnostic suggestions and highlights the symptoms or vitals matched from the suggested diagnosis to the contents of EMR. After physician enters a (list of) diagnosis, CDSS detects whether there is misdiagnosis or missed diagnosis and explains why it is very likely a misdiagnosis. The alert will disappear only when no misdiagnosis or missed diagnosis is detected at all. Then, the treatment recommendation will be given on the correct diagnosis.

Article Snippet: Among the providers of cognitive computing-based CDSS, Baidu, iFLYTEK, and Dr. Mayson are the representatives of the real systems in China.

Techniques: Biomarker Discovery, Diagnostic Assay