Machine learning predicts individual malaria outcomes, disease progression

Machine learning analysis of patient data was able to predict the severity of malaria outcomes and the progression of the disease with significantly greater success than previous approaches, according to the results of a study published July 10 in npj Digital Medicine.

Researchers analyzed the medical data of Gambian children with severe malaria to identify the most significant clinical features that led to death from the disease, and to map out the onset of these symptoms. This analysis was verified — with "remarkable" agreement — by clinicians with experience treating malaria, who could largely offer only educated guesses about the disease's early stages, since their first encounters with malaria patients typically come late in its onset.

A machine learning model trained on that analysis was therefore able to predict the disease progression and outcomes of pediatric malaria on an individual level, compared to the generalized, population-wide assumptions typically made in malaria treatment.

The novel method represents an innovative and viable approach for determining disease pathways by forming predictions based on past patients' data, since it would be unethical to study the onset of a disease in a clinical study without providing treatment. The method therefore "aligns with the goals of precision medicine and makes full use of available biomedical data; we anticipate that it may also find use in numerous other diseases and clinical contexts," the study's authors wrote.

More articles about AI:
Health tech consortium releases ethical guidelines for AI use in healthcare
Geisinger developing AI solutions to detect, prevent high-burden diseases
GlaxoSmithKline poaches another biotech exec for AI team: 3 notes

© Copyright ASC COMMUNICATIONS 2019. Interested in LINKING to or REPRINTING this content? View our policies by clicking here.

 

Top 40 Articles from the Past 6 Months