Machine Learning Models to Predict Cognitive Impairment of Rodents Subjected to Space Radiation

Abstract

This research uses machine-learned computational analyses to predict the cognitive performance impairment of rats induced by irradiation. The experimental data in the analyses is from a rodent model exposed to ≤15 cGy of individual galactic cosmic radiation (GCR) ions: 4He, 16O, 28Si, 48Ti, or 56Fe, expected for a Lunar or Mars mission. This work investigates rats at a subject-based level and uses performance scores taken before irradiation to predict impairment in attentional set-shifting (ATSET) data post-irradiation. Here, the worst performing rats of the control group define the impairment thresholds based on population analyses via cumulative distribution functions, leading to the labeling of impairment for each subject. A significant finding is the exhibition of a dose-dependent increasing probability of impairment for 1 to 10 cGy of 28Si or 56Fe in the simple discrimination (SD) stage of the ATSET, and for 1 to 10 cGy of 56Fe in the compound discrimination (CD) stage. On a subject-based level, implementing machine learning (ML) classifiers such as the Gaussian naïve Bayes, support vector machine, and artificial neural networks identifies rats that have a higher tendency for impairment after GCR exposure. The algorithms employ the experimental prescreen performance scores as multidimensional input features to predict each rodent’s susceptibility to cognitive impairment due to space radiation exposure. The receiver operating characteristic and the precision-recall curves of the ML models show a better prediction of impairment when 56Fe is the ion in question in both SD and CD stages. They, however, do not depict impairment due to 4He in SD and 28Si in CD, suggesting no dose-dependent impairment response in these cases. One key finding of our study is that prescreen performance scores can be used to predict the ATSET performance impairments. This result is significant to crewed space missions as it supports the potential of predicting an astronaut’s impairment in a specific task before spaceflight through the implementation of appropriately trained ML tools. Future research can focus on constructing ML ensemble methods to integrate the findings from the methodologies implemented in this study for more robust predictions of cognitive decrements due to space radiation exposure.

Predicting Space Radiation Single Ion Exposure in Rodents: A Machine Learning Approach

“…Working with subject matter experts to develop a cost matrix (Chawla, 2005) that weighs the relative importance of type I and type II classification errors for GCR ion exposure prediction would help elucidate the extent of the significance in this greater-thanchance classifier performance in application. Similarly, a recent parallel effort by our group (Matar et al, 2021) evaluates the susceptibility to cognitive performance impairment in rodents due to space radiation exposure by demonstrating a capability to predict cognitive performance impairment in individual rodents using their respective pre-irradiation performance scores with ML. The findings demonstrate that prescreen performance scores can be used as features with ML to predict ATSET performance impairments as a direct method of predicting impairment.…”

Human Health during Space Travel: State-of-the-Art Review

“…The Edge TPU and other similar low-power AI ASICs (i.e., Google Coral Edge TPU, NVIDIA Jetson Nano, Intel Neural Compute Stick 2) could advance image segmentation AI models, allowing for these models to be feasibly deployed in ultrasound point-of-care settings (e.g., detection of DVT, structural heart disease, hemodynamic changes), even procedural guidance [100]. AI-based system designed to guide non-physicians on the proper acquisition of medical diagnostic testing using The Edge TPU and deep reinforcement learning AI-enhanced 3D-imaging technology (e.g., micro-CT scanners) [101] Utilization of the Edge TPU and other similar low-power AI ASICs to provide the necessary processing power for high-performance parallel-processing space research [102] Intervention AI-guided minor surgical procedures [such as incision and drainage (I&D)] using next-generation High Performance Spaceflight Computing (HPSC) [103] AI-assisted, remotely controlled robotic PCI and robotic laparoscopic surgery (e.g., telecholecystectomy and teleappendectomy); made possible by a reduction in communication latency beyond a lag of 200 ms [103][104][105][106][107] Using AI predictions of drug metabolism and effectiveness based on an individual's multiomic data prior to medication or supplement distribution (e.g., melatonin, immune supplements, probiotics) [108,109] Disease Prevention AI-integrated space suits (e.g., exoskeletons) to maximize EVA time and operating pressure, and minimize space radiation exposure [110,111] 3D printing of personalized devices (e.g., ear plugs to prevent noise source generated from man-made sources), space shields, space suits for use in emergency scenarios [112][113][114] AI-based chatbots or social media could potentially be used to prevent anxiety and depression during long-duration space travel. The Edge TPU could potentially be used in advancing an internet or social media for the moon, known as LunaNet [115].…”

Quantitative proteomic analytic approaches to identify metabolic changes in the medial prefrontal cortex of rats exposed to space radiation

“…There is a comprehensive body of data on the effect that a wide spectrum of SR species has on performance in the attentional set shifting (ATSET) assay (Parihar et al, 2016;Britten et al, 2018;Britten et al, 2020a;Britten et al, 2021b;Burket et al, 2021). These data sets are now being analyzed with machine learning assisted computational approaches to fully characterize the cognitive deficits induced (Matar et al, 2021;Prelich et al, 2021). However, a readily identifiable consequence of SR exposure is the loss of performance in the Simple Discrimination (SD) stage of the ATSET test.…”