Prefrontal fNIRS-based clinical data analysis of brain functions in individuals abusing different types of drugs

Abstract

Background The activation degree of the orbitofrontal cortex (OFC) functional area in drug abusers is directly related to the craving for drugs and the tolerance to punishment. Currently, among the clinical research on drug rehabilitation, there has been little analysis of the OFC activation in individuals abusing different types of drugs, including heroin, methamphetamine, and mixed drugs. Therefore, it becomes urgently necessary to clinically investigate the abuse of different drugs, so as to explore the effects of different types of drugs on the human brain. Methods Based on prefrontal high-density functional near-infrared spectroscopy (fNIRS), this research designs an experiment that includes resting and drug addiction induction. Hemoglobin concentrations of 30 drug users (10 on methamphetamine, 10 on heroin, and 10 on mixed drugs) were collected using fNIRS and analyzed by combining algorithm and statistics. Results Linear discriminant analysis (LDA), Support vector machine (SVM) and Machine-learning algorithm was implemented to classify different drug abusers. Oxygenated hemoglobin (HbO2) activations in the OFC of different drug abusers were statistically analyzed, and the differences were confirmed. Innovative findings: in both the Right-OFC and Left-OFC areas, methamphetamine abusers had the highest degree of OFC activation, followed by those abusing mixed drugs, and heroin abusers had the lowest. The same result was obtained when OFC activation was investigated without distinguishing the left and right hemispheres. Conclusions The findings confirmed the significant differences among different drug abusers and the patterns of OFC activations, providing a theoretical basis for personalized clinical treatment of drug rehabilitation in the future.

Machine learning with neuroimaging biomarkers: Application in the diagnosis and prediction of drug addiction

“…To be clarified, although diagnosis of drug addiction can be figured out in a routine way like urine test or self-reported drug usage, early diagnostic neuroimaging features lay a foundation to guide clinical treatment and evaluate the therapeutic effect of addiction in the future. 61,110 Besides, neural features are instrumental in observing the treatment response of addicts, which will guide treatment-oriented clinical practice. 113 For example, after combining treatment attendance, the performance of SVM classifier to predict treatment response of cocaine abusers was improved through positron emission tomography (PET) signal of ventral striatum.…”

“…49,124 Moreover, thanks to possibly more than one kind of addictive drug taken by subjects and peculiar brain imaging features created by varied substances, it is indispensable to construct higher-order multi-classification models, consequently improving its application scope, rather than being limited to dichotomous classification. 55,61 In this context, we elucidate the most robust neurobiological predictors engaged in diagnosing clinical status and predicting addiction risk through overviewing the ML methods, thereby learning the current and future pathophysiological status of addicts' neurobiology.…”

“…62,63 Besides, potential neural features extracted by deep learning have poor interpretability in clinical practice. 61,110 Overfitting of ML and deep learning models remains an issue that deserves to be solved urgently. 120 There are still some dilemmas needed to be settled in the future neuroimaging of addiction.…”

“…61 Researchers compared classifying accuracy among linear discriminant analysis, SVM and CNN, and CNN is preferable, 61 ultimately finding that the OFC of MA abusers is the highest activated whereas heroin users were lowest, 61 which lays the foundation for the individualized treatment of rehabilitation. 61…”

“…The architecture of a convolutional neural network (CNN) is a multi‐layered feed‐forward neural network, made by stacking many hidden layers on top of each other in sequence 109 . It is a representative algorithm of deep learning and is designed for processing structured arrays of data such as images, bringing about the extraction of hidden features from medical images and function of diseases' classification 61,110 …”