The Correlation-Base-Selection Algorithm for Diagnostic Schizophrenia Based on Blood-Based Gene Expression Signatures

A primer on the use of machine learning to distil knowledge from data in biological psychiatry

Applications of machine learning in the biomedical sciences are growing rapidly. This growth has been spurred by diverse cross-institutional and interdisciplinary collaborations, public availability of large datasets, an increase in the accessibility of analytic routines, and the availability of powerful computing resources. With this increased access and exposure to machine learning comes a responsibility for education and a deeper understanding of its bases and bounds, borne equally by data scientists seeking to ply their analytic wares in medical research and by biomedical scientists seeking to harness such methods to glean knowledge from data. This article provides an accessible and critical review of machine learning for a biomedically informed audience, as well as its applications in psychiatry. The review covers definitions and expositions of commonly used machine learning methods, and historical trends of their use in psychiatry. We also provide a set of standards, namely Guidelines for REporting Machine Learning Investigations in Neuropsychiatry (GREMLIN), for designing and reporting studies that use machine learning as a primary data-analysis approach. Lastly, we propose the establishment of the Machine Learning in Psychiatry (MLPsych) Consortium, enumerate its objectives, and identify areas of opportunity for future applications of machine learning in biological psychiatry. This review serves as a cautiously optimistic primer on machine learning for those on the precipice as they prepare to dive into the field, either as methodological practitioners or well-informed consumers.

Experimental exposure to gasohol impairs sperm quality with recognition of the classification pattern of exposure groups by machine learning algorithms

Contamination caused by leakage at gas stations leads to possible exposure of the general population when in contact with contaminated water and soil. The present study aimed to evaluate the reproductive effects of exposure of adult male rats to gasohol and evaluate the performance of machine learning (ML) algorithms for pattern recognition and classification of the exposure groups. Rats were orally exposed to 0 (control), 16 (EA), 160 (EB), or 800 mg kg^-1 bw day^-1 of gasohol (EC), for 30 consecutive days. Sperm quality of the groups exposed to two higher doses was reduced in comparison to the control group. The sperm parameters decreased were: daily sperm production, sperm number in the caput/corpus epididymis, progressive motility, mitochondrial activity, and acrosomal membrane integrity. Sperm transit time in the epididymis cauda and sperm isolated head were increased in EB and EC. Sertoli cells number was decreased in these groups, but their support capacity was maintained. ML methods were used to identify patterns between samples of control and exposure groups. The results obtained by ML methods were very promising, obtaining about 90% of accuracy. It was concluded that the exposure of rats to different doses of gasohol impair spermatogenesis and sperm quality, with a recognizable classification pattern of exposure groups at ML.

Evaluation of genotoxic effects in Brazilian agricultural workers exposed to pesticides and cigarette smoke using machine-learning algorithms

Monitoring exposure to xenobiotics by biomarker analyses, such as a micronucleus assay, is extremely important for the precocious detection and prevention of diseases, such as oral cancer. The aim of this study was to evaluate genotoxic effects in rural workers who were exposed to cigarette smoke and/or pesticides and to identify possible classification patterns in the exposure groups. The sample included 120 participants of both sexes aged between 18 and 39, who were divided into the following four groups: control group (CG), smoking group (SG), pesticide group (PG), and smoking + pesticide group (SPG). Their oral mucosa cells were stained with Giemsa for cytogenetic analysis. The total numbers of nuclear abnormalities (CG = 27.16 ± 14.32, SG = 118.23 ± 74.78, PG = 184.23 ± 52.31, and SPG = 191.53 ± 66.94) and micronuclei (CG = 1.46 ± 1.40, SG = 12.20 ± 10.79, PG = 21.60 ± 8.24, and SPG = 20.26 ± 12.76) were higher (p < 0.05) in the three exposed groups compared to the GC. In this study, we considered several different classification algorithms (the artificial neural network, K-nearest neighbors, support vector machine, and optimum path forest). All of the algorithms displayed good classification (accuracy > 80%) when using dataset2 (without the redundant exposure type SPG). It is clear that the data form a robust pattern and that classifiers could be successfully trained on small datasets from the exposure groups. In conclusion, exposing agricultural workers to pesticides and/or tobacco had genotoxic potential, but concomitant exposure to xenobiotics did not lead to additive or potentiating effects.