International Journal of Applied Sciences: Current and Future Research Trends

Caffeine and Parkinson’s Disease: A Comprehensive Review

Muzamil Choudhry — 2022-06-06

Introduction: Parkinson’s disease is one of the most common neurodegenerative disorders, affecting 1% of the population older than 60 years. In search for a novel treatment and prevention of PD in the last two decades, research studies on both humans and mice have discovered the neuro protective effects of caffeine. These studies indicate an inverse relationship between the caffeine and PD which point towards the positive effects of moderate caffeine consumption on Parkinson’s onset and symptoms. Objective: This article explores and analyzes data from the last two decades of published research conducted on both humans and mice to understand the potential molecular mechanisms responsible for the neuro protcetive effects of caffeine , and how it improves motor and non motor deficits related to PD .Results: The neuro protective effects of caffeine have been found to depend on various factors, including sex, caffeine metabolism rate, and smoking. For example , caffeine is beneficial in preventing and slowing down the progression of PD in all tested cohorts of men, but only in varying cohorts of women, owing to the competition between estrogen and caffeine for metabolism via CYP1A2. The cohort with the least positive outcome for caffeine was post-menopausal women who were actively taking hormone-replacement therapy (HRT). Similarly, variation in the rate of caffeine metabolism from individual to individual and smoking induce an increase in hepatic enzymes affect the caffeine out come in Parkinson disease (PD). Conclusion: Due to the complexity of biological mechanisms and the methodology of scientific research, it is difficult to fully elucidate the benefits of caffeine on PD development and progression despite substantial evidence indicating a lower incidence of PD among caffeine users.

Stationary Distribution of a Mathematical Model

Lahcen Boulaasair — 2022-06-23

This paper aims to study a new class of a stochastic SIS (susceptible, infected, susceptible) epidemic model where the transmission coefficient of the disease and the death rate are perturbed. By using the Khasminski theory, we establish suitable conditions under which the stochastic SIS model has a unique stationary distribution. The stationary of such model means that the disease will prevail.

Mammography Datasets for Neural Networks - Survey

Adam Mračko — 2022-06-29

Recently, deep neural networks have become popular in the mammography field. Data are an integral part of the model in the training phase because training algorithms require a large amount of data to capture the general relation between the model input and its output. The most accessible way of obtaining mammography data for training neural networks is through open-access databases. Our work focuses on a thorough survey of the mammography databases which contain images with a defined abnormality area of interest. The survey involves databases INbreast, Curated Breast Imaging Subset of Digital Database for Screening Mammograph (CBIS-DDSM), and The Mammographic Image Analysis Society Digital Mammogram Database (MIAS). Also, a description of the annotation process of mammography images is included for a better understanding of information gained from datasets.

Survey of Recent Deep Neural Networks with Strong Annotated Supervision in Histopathology

Dominika Petríková — 2022-07-02

Deep learning (DL) and convolutional neural networks (CNNs) have achieved state-of-the-art performance in many medical image analysis tasks. Histopathological images contain valuable information that can be used to make disease diagnosis and create treatment plans. Therefore, application of DL for classification of histological images is rapidly expanding field of research. In this paper, we reviewed recent DL-based classification studies in histopathology using strongly annotated data. We divided them according to training approach and model construction into three categories: fine-tuning or training networks from scratch to make one-stage classification and multi-stage classification. Papers summarized in this study are covering wide area of application (e.g., breast, lung, colon, brain, kidney).