Machine learning for predicting cognitive deficits using auditory and demographic factors.

IMPORTANCE: Predicting neurocognitive deficits using complex auditory assessments could change how cognitive dysfunction is identified, and monitored over time. Detecting cognitive impairment in people living with HIV (PLWH) is important for early intervention, especially in low- to middle-income countries where most cases exist. Auditory tests relate to neurocognitive test results, but the incremental predictive capability beyond demographic factors is unknown. OBJECTIVE: Use machine learning to predict neurocognitive deficits, using auditory tests and demographic factors. SETTING: The Infectious Disease Center in Dar es Salaam, Tanzania. PARTICIPANTS: Participants were 939 Tanzanian individuals from Dar es Salaam living with and without HIV who were part of a longitudinal study. Patients who had only one visit, a positive history of ear drainage, concussion, significant noise or chemical exposure, neurological disease, mental illness, or exposure to ototoxic antibiotics (e.g., gentamycin), or chemotherapy were excluded. This provided 478 participants (349 PLWH, 129 HIV-negative). Participant data were randomized to training and test sets for machine learning. MAIN OUTCOME(S) AND MEASURE(S): The main outcome was whether auditory variables combined with relevant demographic variables could predict neurocognitive dysfunction (defined as a score of <26 on the Kiswahili Montreal Cognitive Assessment) better than demographic factors alone. The performance of predictive machine learning algorithms was primarily evaluated using the area under the receiver operational characteristic curve. Secondary metrics for evaluation included F1 scores, accuracies, and the Youden's indices for the algorithms. RESULTS: The percentage of individuals with cognitive deficits was 36.2% (139 PLWH and 34 HIV-negative). The Gaussian and kernel naïve Bayes classifiers were the most predictive algorithms for neurocognitive impairment. Algorithms trained with auditory variables had average area under the curve values of 0.91 and 0.87, F1 scores (metric for precision and recall) of 0.81 and 0.76, and average accuracies of 86.3% and 81.9% respectively. Algorithms trained without auditory variables as features were statistically worse (p < .001) in both the primary measure of area under the curve (0.82/0.78) and the secondary measure of accuracy (72.3%/74.5%) for the Gaussian and kernel algorithms respectively. CONCLUSIONS AND RELEVANCE: Auditory variables improved the prediction of cognitive function. Since auditory tests are easy-to-administer and often naturalistic tasks, they may offer objective measures or predictors of neurocognitive performance suitable for many global settings. Further research and development into using machine learning algorithms for predicting cognitive outcomes should be pursued.

Central auditory test performance predicts future neurocognitive function in children living with and without HIV.

Tests of the brain's ability to process complex sounds (central auditory tests) correlate with overall measures of neurocognitive performance. In the low- middle-income countries where resources to conduct detailed cognitive testing is limited, tests that assess the central auditory system may provide a novel and useful way to track neurocognitive performance. This could be particularly useful for children living with HIV (CLWH). To evaluate this, we administered central auditory tests to CLWH and children living without HIV and examined whether central auditory tests given early in a child's life could predict later neurocognitive performance. We used a machine learning technique to incorporate factors known to affect performance on neurocognitive tests, such as education. The results show that central auditory tests are useful predictors of neurocognitive performance and perform as well or in some cases better than factors such as education. Central auditory tests may offer an objective way to track neurocognitive performance in CLWH.

Auditory neural processing in children living with HIV uncovers underlying central nervous system dysfunction.

OBJECTIVE: Central nervous system (CNS) damage from HIV infection or treatment can lead to developmental delays and poor educational outcomes in children living with HIV (CLWH). Early markers of central nervous system dysfunction are needed to target interventions and prevent life-long disability. The frequency following response (FFR) is an auditory electrophysiology test that can reflect the health of the central nervous system. In this study, we explore whether the FFR reveals auditory central nervous system dysfunction in CLWH. STUDY DESIGN: Cross-sectional analysis of an ongoing cohort study. Data were from the child's first visit in the study. SETTING: The infectious disease center in Dar es Salaam, Tanzania. METHODS: We collected the FFR from 151 CLWH and 151 HIV-negative children. To evoke the FFR, three speech syllabi (/da/, /ba/, /ga/) were played monaurally to the child's right ear. Response measures included neural timing (peak latencies), strength of frequency encoding (fundamental frequency and first formant amplitude), encoding consistency (inter-response consistency), and encoding precision (stimulus-to-response correlation). RESULTS: CLWH showed smaller first formant amplitudes ( P  < 0.0001), weaker inter-response consistencies ( P  < 0.0001) and smaller stimulus to response correlations ( P  < 0.0001) than FFRs from HIV-negative children. These findings generalized across the three speech stimuli with moderately strong effect sizes (partial η2 ranged from 0.061 to 0.094). CONCLUSION: The FFR shows auditory central nervous system dysfunction in CLWH. Neural encoding of auditory stimuli was less robust, more variable, and less accurate. As the FFR is a passive and objective test, it may offer an effective way to assess and detect central nervous system function in CLWH.

Sustained changes in digestive physiology and microbiome across sequential generations of zebrafish fed different diets.

Alterations to ratios of protein and fiber in an organism's diet have been shown to structurally and functionally alter its individual digestive physiology. However, it is unclear how these dietary changes may affect phenotypic changes across generations. We utilized feeding trials, morphological analyses, enzyme activities, and 16S rRNA sequencing of the gut microbiome of zebrafish (Danio rerio) to determine how variations to fiber and protein concentrations, kept consistent across sequential generations, affect phenotypic changes. Our results show that Parental (P) and first generation (F1) fish did not differ from each other in terms of their intestine length, intestine mass, enzyme activity levels, and microbial community composition for any of the three experimental diets (high-protein/low-fiber, moderate-protein/fiber, and low-protein/high-fiber). However, each of the three experimental diets for the P and F1 fish, as well as the ancestral diet fish, did have distinct microbial community structure from one another. This indicates that there is a strong dietary effect on digestive physiology and gut microbial community and that these effects are consistent when the diet is kept homogenous across generations.

Shark spiral intestines may operate as Tesla valves.

Looking to nature for inspiration has led to many diverse technological advances. The spiral valve intestine of sharks has provided the opportunity to observe the efficiency of different valve systems. It is supposed that the spiral intestine present in sharks, skates and rays slows the transit rate of digesta through the gut and provides increased surface area for the absorption of nutrients. In this investigation, we use a novel technique-creating three-dimensional reconstructions from CT scans of spiral intestines-to describe the morphology of the spiral intestine of at least one species from 22 different shark families. We discuss the morphological data in an evolutionary, dietary and functional context. The evolutionary analyses suggest that the columnar morphology is the ancestral form of the spiral intestine. Dietary analyses reveal no correlation between diet type and spiral intestine morphology. Flow rate was slowed significantly more when the two funnel-shaped spiral intestines were subjected to flow in the posterior to anterior direction, indicating their success at producing unidirectional flow, similar to a Tesla valve. These data are available to generate additional three-dimensional morphometrics, create computational models of the intestine, as well as to further explore the function of the gastrointestinal tract of sharks in structural and physiological contexts.

Three-dimensional movements of the pectoral fin during yaw turns in the Pacific spiny dogfish, Squalus suckleyi.

Fish pectoral fins move in complex ways, acting as control surfaces to affect force balance during swimming and maneuvering. Though objectively less dynamic than their actinopterygian relatives, shark pectoral fins undergo complex conformational changes and movements during maneuvering. Asynchronous pectoral fin movement is documented during yaw turning in at least two shark species but the three-dimensional (3D) rotation of the fin about the body axes is unknown. We quantify the 3D actuation of the pectoral fin base relative to the body axes. We hypothesized that Pacific spiny dogfish rotate pectoral fins with three degrees of freedom relative to the body during volitional turning. The pectoral fin on the inside of the turn is consistently protracted, supinated and depressed. Additionally, turning angular velocity increased with increasing fin rotation. Estimated drag on the fin increased and the shark decelerated during turning. Based on these findings, we propose that Pacific spiny dogfish uses drag-based turning during volitional swimming. Post-mortem muscle stimulation revealed depression, protraction and supination of the pectoral fin through stimulation of the ventral and cranial pterygoideus muscles. These data confirm functional hypotheses about pectoral fin musculature and suggest that Pacific spiny dogfish actively rotate pectoral fins to facilitate drag-based turning.This article has an associated First Person interview with the first author of the paper.

Seagrass digestion by a notorious 'carnivore'.

What an animal consumes and what an animal digests and assimilates for energetic demands are not always synonymous. Sharks, uniformly accepted as carnivores, have guts that are presumed to be well suited for a high-protein diet. However, the bonnethead shark (Sphyrna tiburo), which is abundant in critical seagrass habitats, has been previously shown to consume copious amounts of seagrass (up to 62.1% of gut content mass), although it is unknown if they can digest and assimilate seagrass nutrients. To determine if bonnetheads digest seagrass nutrients, captive sharks were fed a 13C-labelled seagrass diet. Digestibility analyses, digestive enzyme assays and stable isotope analyses were used to determine the bonnethead shark's capacity for digesting and assimilating seagrass material. Compound-specific stable isotope analysis showed that sharks assimilated seagrass carbon (13.6 ± 6.77‰ δ13C mean ± s.d. for all sharks and all amino acid types analysed) with 50 ± 2% digestibility of seagrass organic matter. Additionally, cellulose-component-degrading enzyme activities were detected in shark hindguts. We show that a coastal shark is digesting seagrass with at least moderate efficiency, which has ecological implications due to the stabilizing role of omnivory and nutrient transport within fragile seagrass ecosystems.

The effects of protein and fiber content on gut structure and function in zebrafish (Danio rerio).

Chemical reactor theory (CRT) suggests that the digestive tract functions as a chemical reactor for processing food. Presumably, gut structure and function should match diet to ensure adequate nutrient and energy uptake to maintain performance. Within CRT, dietary biochemical composition is the most important factor affecting gut structure and function in vertebrates. We fed Danio rerio (zebrafish) diets ranging from high- to moderate- to low-quality (i.e., ranging from high-protein, low-fiber to low-protein, high-fiber), and observed how gut length and surface area, as well as the activity levels of digestive enzymes (amylase, maltase, trypsin, aminopeptidase, and lipase) shifted in response to these dietary changes. Fish on the low-quality diet had the longest guts with the largest intestinal epithelial surface area and enterocyte cellular volumes. Fish on the moderate-quality diet had intermediate values of most of these parameters, and fish on the high-quality diet, the lowest. These data largely support CRT. Digestive enzyme activity levels were generally elevated in fish fed the moderate- and low-quality diets, but were highest in the fish fed the moderate-quality diet, suggesting that a diet with protein levels closest to that of the natural diet of D. rerio (they are omnivorous in nature) may elicit the best gut performance. However, fish fed the carnivore diet reached the largest terminal body size. Our results support CRT in terms of gut structure; however, our enzyme results do not necessarily agree with CRT and largely depend on which enzyme is discussed. In particular, the evidence for lipase activities being elevated in the fish fed the low-protein, high-fiber diet perhaps reflects a lipid-scavenging mechanism in fish consuming high-fiber foods rather than CRT.