Prediction models in the design of neural network based ECG classifiers: a neural network and genetic programming approach

Antiretroviral therapy non-adherence and its relationship with cognitive impairment, alcohol use disorder, and depression in adolescents living with HIV

OBJECTIVE

We aimed to compare antiretroviral non-adherence in the behaviourally infected (BIAs) and congenitally infected adolescents (CIAs) and explore its associations with depression, cognitive impairment, and alcohol use disorder (AUD) in adolescents living with HIV(ALWHIV) in Botswana.

METHODS

This study was a cross-sectional, comparative, multi-center research that involved collecting samples from different HIV clinics in Botswana. Of the 622 ALWHIV, 223 were identified as BIAs and 399 as CIAs. They were evaluated using various tools such as MINI-KID for psychiatric disorders, DSM-5 for AUD, CAT-rapid for cognitive assessment, and Visual Analogue Scale (VAS) for non-adherence (the outcome). The data were analysed using both bivariate and multivariate regression analyses.

RESULTS

The participants' mean age (SD) was 17.7(1.60). The CIAs were more likely to have cognitive impairment (t -7.25; p < 0.01), while the BIAs had more depression (χ2 = 5.86; p = 0.016) and AUD (χ2 = 4.39; p = 0.036) and were more likely to be non-adherent (t = 3.14; p = 0.002). In the CIA group, cognitive impairment (AOR = 2.86; 95% CI:1.77-4.64) (AOR = 2.79; 95%CI:1.73-4.48) and depression (AOR = 2.69; 95%CI:1.48-4.90 were associated with ART non-adherence. In the BIA group, depression (AOR = 2.55; 95%CI:1.27-5.16), AUD (AOR = 2.58; 95%CI:1.21-5.49) and struggling to accept status (AOR = 2.54; 95%CI:1.41-4.56) predicted non-adherence to treatment.

CONCLUSION

The two groups of adolescents differ regarding ART non-adherence and associated psychosocial issues, indicating the need for differentiated care to address non-adherence in the ALWHIV, especially in high-burden, resource-constrained settings, such as Botswana.

Smart ECG Biosensor Design with an Improved ANN Performance Based on the Taguchi Optimizer

This paper aims to design a smart biosensor to predict electrocardiogram (ECG) signals in a specific auscultation site from other ECG signals measured from other measurement sites. The proposed design is based on a hybrid architecture using the Artificial Neural Networks (ANNs) model and Taguchi optimizer to avoid the ANN issues related to hyperparameters and to improve its accuracy. The proposed approach aims to optimize the number and type of inputs to be considered for the ANN model. Indeed, different combinations are considered in order to find the optimal input combination for the best prediction quality. By identifying the factors that influence a model’s prediction and their degree of importance via the modified Taguchi optimizer, the developed biosensor improves the prediction accuracy of ECG signals collected from different auscultation sites compared to the ANN-based biosensor. Based on an actual database, the simulation results show that this improvement is significant; it can reach more than 94% accuracy.

Evaluation of inherent performance of intelligent medical decision support systems: utilising neural networks as an example

Researchers who design intelligent systems for medical decision support, are aware of the need for response to real clinical issues, in particular the need to address the specific ethical problems that the medical domain has in using black boxes. This means such intelligent systems have to be thoroughly evaluated, for acceptability. Attempts at compliance, however, are hampered by lack of guidelines. This paper addresses the issue of inherent performance evaluation, which researchers have addressed in part, but a Medline search, using neural networks as an example of intelligent systems, indicated that only about 12.5% evaluated inherent performance adequately. This paper aims to address this issue by concentrating on the possible evaluation methodology, giving a framework and specific suggestions for each type of classification problem. This should allow the developers of intelligent systems to produce evidence of a sufficiency of output performance evaluation.