Predicting Fetal Alcohol Spectrum Disorders Using Machine Learning Techniques: Multisite Retrospective Cohort Study.

BACKGROUND: Fetal alcohol syndrome (FAS) is a lifelong developmental disability that occurs among individuals with prenatal alcohol exposure (PAE). With improved prediction models, FAS can be diagnosed or treated early, if not completely prevented. OBJECTIVE: In this study, we sought to compare different machine learning algorithms and their FAS predictive performance among women who consumed alcohol during pregnancy. We also aimed to identify which variables (eg, timing of exposure to alcohol during pregnancy and type of alcohol consumed) were most influential in generating an accurate model. METHODS: Data from the collaborative initiative on fetal alcohol spectrum disorders from 2007 to 2017 were used to gather information about 595 women who consumed alcohol during pregnancy at 5 hospital sites around the United States. To obtain information about PAE, questionnaires or in-person interviews, as well as reviews of medical, legal, or social service records were used to gather information about alcohol consumption. Four different machine learning algorithms (logistic regression, XGBoost, light gradient-boosting machine, and CatBoost) were trained to predict the prevalence of FAS at birth, and model performance was measured by analyzing the area under the receiver operating characteristics curve (AUROC). Of the total cases, 80% were randomly selected for training, while 20% remained as test data sets for predicting FAS. Feature importance was also analyzed using Shapley values for the best-performing algorithm. RESULTS: Overall, there were 20 cases of FAS within a total population of 595 individuals with PAE. Most of the drinking occurred in the first trimester only (n=491) or throughout all 3 trimesters (n=95); however, there were also reports of drinking in the first and second trimesters only (n=8), and 1 case of drinking in the third trimester only (n=1). The CatBoost method delivered the best performance in terms of AUROC (0.92) and area under the precision-recall curve (AUPRC 0.51), followed by the logistic regression method (AUROC 0.90; AUPRC 0.59), the light gradient-boosting machine (AUROC 0.89; AUPRC 0.52), and XGBoost (AUROC 0.86; AURPC 0.45). Shapley values in the CatBoost model revealed that 12 variables were considered important in FAS prediction, with drinking throughout all 3 trimesters of pregnancy, maternal age, race, and type of alcoholic beverage consumed (eg, beer, wine, or liquor) scoring highly in overall feature importance. For most predictive measures, the best performance was obtained by the CatBoost algorithm, with an AUROC of 0.92, precision of 0.50, specificity of 0.29, F1 score of 0.29, and accuracy of 0.96. CONCLUSIONS: Machine learning algorithms were able to identify FAS risk with a prediction performance higher than that of previous models among pregnant drinkers. For small training sets, which are common with FAS, boosting mechanisms like CatBoost may help alleviate certain problems associated with data imbalances and difficulties in optimization or generalization.

Improving brain B0 shimming using an easy and accessible multi-coil shim array at ultra-high field.

OBJECT: Improve shimming capabilities of ultra-high field systems, with addition of an accessible low-complexity B0 shim array for head MRI at 7 T. MATERIALS AND METHODS: An eight channel B0 shim coil array was designed as a tradeoff between shimming improvement and construction complexity, to provide an easy to use shim array that can be employed with the standard 7 T head coil. The array was interfaced using an open-source eight-channel shim amplifier rack. Improvements in field homogeneity for whole-brain and slice-based shimming were compared to standard second-order shimming, and to more complex higher order dynamic shimming and shim arrays with 32 and 48 channels. RESULTS: The eight-channel shim array provided 12% improvement in whole brain static shimming and provided 33% improvement when using slice-based shimming. With this, the eight-channel array performed similar to third-order dynamic shimming (without the need for higher order eddy current compensation). More complex shim arrays with 32 and 48 channels performed better, but require a dedicated RF coil. DISCUSSION: The designed eight-channel shim array provides a low-complexity and low-cost approach for improving B0 field shimming on an ultra-high field system. In both static and dynamic shimming, it provides improved B0 homogeneity over standard shimming.

Antimicrobial properties of tomato leaves, stems, and fruit and their relationship to chemical composition.

BACKGROUND: We previously reported that the tomato glycoalkaloid tomatine inhibited the growth of Trichomonas vaginalis strain G3, Tritrichomonas foetus strain D1, and Tritrichomonas foetus-like strain C1 that cause disease in humans and farm and domesticated animals. The increasing prevalence of antibiotic resistance requires development of new tools to enhance or replace medicinal antibiotics. METHODS: Wild tomato plants were harvested and divided into leaves, stems, and fruit of different colors: green, yellow, and red. Samples were freeze dried and ground with a handheld mill. The resulting powders were evaluated for their potential anti-microbial effects on protozoan parasites, bacteria, and fungi. A concentration of 0.02% (w/v) was used for the inhibition of protozoan parasites. A high concentration of 10% (w/v) solution was tested for bacteria and fungi as an initial screen to evaluate potential anti-microbial activity and results using this high concentration limits its clinical relevance. RESULTS: Natural powders derived from various parts of tomato plants were all effective in inhibiting the growth of the three trichomonads to varying degrees. Test samples from leaves, stems, and immature 'green' tomato peels and fruit, all containing tomatine, were more effective as an inhibitor of the D1 strain than those prepared from yellow and red tomato peels which lack tomatine. Chlorogenic acid and quercetin glycosides were present in all parts of the plant and fruit, while caffeic acid was only found in the fruit peels. Any correlation between plant components and inhibition of the G3 and C1 strains was not apparent, although all the powders were variably effective. Tomato leaf was the most effective powder in all strains, and was also the highest in tomatine. S. enterica showed a minor susceptibility while B. cereus and C. albicans fungi both showed a significant growth inhibition with some of the test powders. The powders inhibited growth of the pathogens without affecting beneficial lactobacilli found in the normal flora of the vagina. CONCLUSIONS: The results suggest that powders prepared from tomato leaves, stems, and green tomato peels and to a lesser extent from peels from yellow and red tomatoes offer potential multiple health benefits against infections caused by pathogenic protozoa, bacteria, and fungi, without affecting beneficial lactobacilli that also reside in the normal flora of the vagina.

Anti-Parasitic Activity of Cherry Tomato Peel Powders.

Trichomoniasis in humans, caused by the protozoal parasite Trichomonas vaginalis, is the most common non-viral sexually transmitted disease, while Tritrichomonas foetus causes trichomonosis, an infection of the gastrointestinal tract and diarrhea in farm animals and domesticated cats. As part of an effort to determine the inhibitory effects of plant-based extracts and pure compounds, seven commercially available cherry tomato varieties were hand-peeled, freeze-dried, and pounded into powders. The anti-trichomonad inhibitory activities of these peel powders at 0.02% concentration determined using an in vitro cell assay varied widely from 0.0% to 66.7% against T. vaginalis G3 (human); from 0.9% to 66.8% for T. foetus C1 (feline); and from 0.0% to 81.3% for T. foetus D1 (bovine). The organic Solanum lycopersicum var. cerasiforme (D) peels were the most active against all three trichomonads, inhibiting 52.2% (G3), 66.8% (C1), and 81.3% (D1). Additional assays showed that none of the powders inhibited the growth of foodborne pathogenic bacteria, pathogenic fungi, or non-pathogenic lactobacilli. Tomato peel and pomace powders with high content of described biologically active compounds could serve as functional food and feed additives that might help overcome adverse effects of wide-ranging diseases and complement the treatment of parasites with the anti-trichomonad drug metronidazole.

A 16-channel AC/DC array coil for anesthetized monkey whole-brain imaging at 7T.

Functional magnetic resonance imaging (fMRI) in monkeys is important for bridging the gap between invasive animal brain studies and non-invasive human brain studies. To resolve the finer functional structure of the monkey brain, ultra-high-field (UHF) MR is essential, and high-performance, close-fitting RF receive coils are typically desired to fully leverage the intrinsic gains provided by UHF MRI. Moreover, static field (B0) inhomogeneity arising from the tissue susceptibility interface is more severe at UHF, presenting an obstacle to achieving high-resolution fMRI. B0 shim of the monkey head is challenging due to its smaller size and more complex sources of B0 offsets in multi-modal imaging tasks. In the present work, we have customized an array coil for lightly-anesthetized monkey fMRI in the 7T human scanner that combines RF and multi-coil (MC) B0 shim functionality (also referred to as AC/DC coils) to provide high imaging SNR and high-spatial-order, rapidly switchable B0-shim capability. Additional space was retained on the coil to render it compatible with monkey multi-modal imaging studies. Both MC global (whole-volume) and dynamic (slice-optimized) shim methods were tested and evaluated, and the benefits of MC shim for fMRI experiments was also studied. A minor reduction in RF coil performance was found after introducing additional B0 shim circuitry. However, the proposed RF coil provided higher image SNR and more uniform contrast compared to a commercially available coil for human knee imaging. Compared with static 2nd-order shim, the B0 inhomogeneity was reduced by 56.8%, and 95-percentile B0 offset was reduced to within 28.2 Hz through MC shim, versus 68.7 Hz with 2nd-order static shim. As a result, functional image quality could be improved, and brain activation can be better detected using the proposed AC/DC monkey coil.

Long-Term Modeling and Monitoring of Neuromusculoskeletal System Performance Using Tattoo-Like EMG Sensors.

This paper introduces stretchable, long-term wearable, tattoo-like dry surface electrodes for highly repeatable electromyography (EMG). The tattoo-like sensors are hair thin, skin compliant and can be laminated on human skin just like a temporary transfer tattoo, which enables multi-day noninvasive but intimate contact with the skin even under severe skin deformation. The new electrodes were used to facilitate a system-based approach to tracking of long-term fatiguing and recovery processes in a human neuromusculoskeletal (NMS) system, which was based on establishing an autoregressive moving average model with exogenous inputs (ARMAX model) relating signatures extracted from the surface electromyogram (sEMG) signals collected using the tattoo-like sensors, and the corresponding hand grip force (HGF) serving as the model output. Performance degradation of the relevant NMS system was evaluated by tracking the evolution of the errors of the ARMAX model established using the data corresponding to the rested (fresh) state of any given subject. Results from several exercise sessions clearly showed repeated patterns of fatiguing and resting, with a notable point that these patterns could now be quantified via dynamic models relating the relevant muscle signatures and NMS outputs.

Assessment of Dry Epidermal Electrodes for Long-Term Electromyography Measurements.

Commercially available electrodes can only provide quality surface electromyography (sEMG) measurements for a limited duration due to user discomfort and signal degradation, but in many applications, collecting sEMG data for a full day or longer is desirable to enhance clinical care. Few studies for long-term sEMG have assessed signal quality of electrodes using clinically relevant tests. The goal of this research was to evaluate flexible, gold-based epidermal sensor system (ESS) electrodes for long-term sEMG recordings. We collected sEMG and impedance data from eight subjects from ESS and standard clinical electrodes on upper extremity muscles during maximum voluntary isometric contraction tests, dynamic range of motion tests, the Jebsen Taylor Hand Function Test, and the Box & Block Test. Four additional subjects were recruited to test the stability of ESS signals over four days. Signals from the ESS and traditional electrodes were strongly correlated across tasks. Measures of signal quality, such as signal-to-noise ratio and signal-to-motion ratio, were also similar for both electrodes. Over the four-day trial, no significant decrease in signal quality was observed in the ESS electrodes, suggesting that thin, flexible electrodes may provide a robust tool that does not inhibit movement or irritate the skin for long-term measurements of muscle activity in rehabilitation and other applications.

NFC-enabled, tattoo-like stretchable biosensor manufactured by "cut-and-paste" method.

The wearables industry is lacking in devices that have the ability to provide valuable biometrics data in a soft, wireless and disposable system. Such a system should be high performance, multifunctional, but battery-free and low cost. Near field communication (NFC) is a wireless communication protocol built in many smartphones nowadays that can read data from battery-free passive tags. As a result, NFC-enabled wearable biosensors have been reported, but they are either unstretchable or have to be manufactured by labor- and time-intensive photolithography and transfer-printing processes. Using a dry and freeform "cut-and-paste" method, we have built a wireless and low-cost stretchable biosensor that integrates temperature sensor, light source/sensor, NFC chip, and antenna. It is battery-free and can be laminated on any part of human skin like a temporary transfer tattoo. The sensor can fully follow the stretching and compression of skin without mechanical failure or delamination. Thus, it is imperceptible to wear and can perform high-fidelity sensing. Potential applications include, but are not limited to, skin thermography and photometry.