A machine learning approach for early prediction of positive blood cultures in neutropenic patients using medical history and hematological parameters
Cancer patients who undergo multiple chemotherapy sessions exhibit a decrease in peripheral neutrophils, which leads to immunodeficiency. Internal infections that affect these patients and are left untreated cause neutropenia. Neutropenia is a life-threatening condition that is a major cause of cancer chemotherapy-related mortality and is caused by a violation of the first line of defense against microorganisms, consisting of a decrease in the absolute number of neutrophils. It will be. Cancer patients with neutropenic sepsis are usually treated with broad-spectrum antibiotics (BSA) as a first-line drug to destroy common bacteria and microorganisms. Broad-spectrum antibiotics lead to antimicrobial resistance (AMR), which subsequently reduces the effectiveness of even specific antibiotics. Doctors rely on her use of BSA because it usually takes two to five days for blood culture results to be available. In this study, we hypothesized that machine learning using hematological parameters would allow early prediction of the presence of bacterial growth in blood cultures and its characterization prior to culture results.
Details of proposed future work
- Predict the class of microorganisms in positive blood cultures, including Gram-negative bacteria, Gram-positive bacteria, and Saccharomyces cerevisiae.
- A scoring system that uses blood test values and biomarkers to determine culture positivity.
