Selecting the optimal activation function for training deep neural networks has always been challenging because it significantly impacts the neural network’s performance and training speed. At this point, researchers are more likely to employ RELU than other well-known activation functions. After RELU, researchers have proposed many activation functions to improve RELU. None of them was capable of surpassing RELU as their most significant rival. SWISH outperformed RELU in several challenging experiments like classification, object detection, and tracking. Replacing RELU units with SWISH, which improves performance in many tasks. This paper proposed a new activation function surpassing Google’s brain’s SWISH function, Which we named AIF. Experiments indicate that our proposed activation function outperforms SWISH in various tasks.

Background: Familial Hypercholesterolemia (FH) is a genetic disorder in lipoprotein metabolism caused by mutations that increase LDL and total cholesterol levels. High LDL and cholesterol levels increase atherosclerosis risk. FH mutations impact the LDL receptor (LDLR) gene, apolipoprotein B, and PCSK9. About 20% of FH cases have a polygenic basis that affects LDL levels. We decided to conduct a systematic review of the available research in this field to provide a thorough genes/proteins network meta-analysis on the impact of drug combinations on the management of heterozygous Familial Hypercholesterolemia (HeFH). This paper reviews and analyzes the literature on the effects of medication combinations on HeFH management. This study investigates articles that analyzed the management and adjuvants of HeFH to recommend forceful drug combinations. Methods: This systematic review and network meta-analysis analyzed the Science Direct, Embase, Scopus, PubMed, Web of Science (ISI), and Google Scholar databases without a lower time limit and up to July 2022. The current study consists of three fundamental stages. Firstly, drug combinations are recommended by reinforcement learning. In the second stage, we used a systematic review to analyze RL's outcomes in diverse populations (with a variety of ages, sex, etc.). Natural Language Processing (NLP) employs context to search these articles. We contrasted manual and NLP-based searches and discovered that NLP could find articles based on MeSH, not simply words. In stage three, we analyze RL outcomes using network meta-analysis. Results: This study uses the RAIN method to investigate the most effective medication combination for managing Heterozygous Familial Hypercholesterolemia (HeFH). Results from the method indicate that the best-recommended scenario is 2.7 times more efficient than the prescription of Ezetimibe as the initial scenario. Conclusion: Our systematic review and network meta-analysis review indicate that a drug combination of Ezetimibe, Pravastatin, and Simvastatin is highly effective. However, additional high-quality clinical trials are required to determine the efficacy and safety of other treatments.

Alzheimer's disease is a neurological illness that worsens gradually. It is known as the most frequent cause of dementia. Moreover, Alzheimer's disease is a general term for memory loss and other cognitive impairments severe enough to disrupt daily life. The disease often grows in middle-aged and elderly people with the gradual loss of cognitive ability. Since the diagnosis of the illness is time consuming, a novel machine learning approach is proposed in this research to diagnose Alzheimer's levels based on Magnetic Resonance Imaging (MRI). More specifically, to provide distinct information for classification we have implemented the proposed method on a native MRI scans dataset containing three MRI channels; namely T1, T2, and, T2 tirm, which are shown in red, blue and green, respectively. The five middle layers of these three types of MRI scans are combined to form five RGB layers. In addition, we use a Convolutional Neural Network to classify each composed layer, and the final disease type is predicted using majority voting. finally, the F1-Score has been used to assess method performance, indicated that 92% of predictions of the proposed model were accurate.

When training deep neural networks tasks, the most popular choices are cross-entropy loss. On the other hand, in general speaking, a decent loss function can take on shapes that are considerably more flexible and ought to be adapted for different activities and datasets. In most of the classification tasks, generally if the true class is not correctly recognized by the network (top1), that class is placed among the five classes with the highest probability (top5). This shows that the network does not necessarily recognize the correct class with a low probability, but a class similar to it (such as 3 vs. 8 in mnist) assigns a higher probability and this causes a mistake in that task. Accordingly, we proposed a loss function deals with the error of class that the neural network incorrectly recognized as correct, in addition to the correct class error. We call our proposed loss as False Positive Loss (FPL), with the intention of viewing and designing loss functions not only through the utilization of true class but also through the utilization of the value of false positive classes. One of the core properties of our proposed loss is full adaptability, which makes False Positive Loss be fully capable of getting reformulated by using other widely used loss functions formulas based on the task or the need of the users. Extensive experimental results demonstrate that our suggested loss function outperforms other well-known losses on a variety of tasks and datasets. As can be observed, the performance of our False Positive Loss is superior to that of the cross-entropy loss when it comes to tasks involving 2D picture classification. We have compared our loss with cross entropy as the most common classification loss function on some models (such as ResNet-18, ResNet-50 and Efficientnet-V2) through classification known as a basic computer vision task. with both random or pre-trained initial weights. As a result, in some cases the models with our loss outperform the same tasks with cross entropy from the viewpoint of some metric (i.e. accuracy and FP). For example, the resnet-50 on cifar-10 dataset with random initialization indicated a top1 accuracy of 94.93 with cross entropy and 95.25 with our loss, while for top5 accuracy the results are 99.86 and 99.87, respectively.

Background Adenocarcinoma of the lung is the most common type of lung cancer, and it is characterized by distinct cellular and molecular features. It occurs when abnormal lung cells multiply out of control and form a tumor in the outer region of the lungs. Adenocarcinoma of the lung is a serious and life-threatening condition that requires effective and timely management to improve the survival and quality of life of the patients. One of the challenges in this cancer treatment is finding the optimal combination of drugs that can target the genes or proteins that are involved in the disease process. Method In this article, we propose a novel method to recommend combinations of trending drugs to target its associated proteins/genes, using a Graph Neural Network (GNN) under the RAIN protocol. The RAIN protocol is a three-step framework that consists of: 1) Applying graph neural networks to recommend drug combinations by passing messages between trending drugs for managing disease and genes that act as potential targets for disease; 2) Retrieving relevant articles with clinical trials that include those proposed drugs in previous step using Natural Language Processing (NLP). The search queries include “Adenocarcinoma of the lung”, “Gefitinib”, “Paclitaxel”, “Icotinib” that searched context based in databases using NLP; 3) Analyzing the network meta-analysis to measure the comparative efficacy of the drug combinations. Result We applied our method to a dataset of nodes and edges that represent the network, where each node is a drug or a gene, and each edge is a p-value between them. We found that the graph neural network recommends combining Gefitinib, Paclitaxel, and Icotinib as the most effective drug combination to target this cancer associated proteins/genes. We reviewed the clinical trials and expert opinions on these medications and found that they support our claim. The network meta-analysis also confirmed the effectiveness of these drugs on associated genes. Conclusion Our method is a novel and promising approach to recommend trending drugs combination to target cancer associated proteins/genes, using graph neural networks under the RAIN protocol. It can help clinicians and researchers to find the best treatment options for patients, and also provide insights into the underlying mechanisms of the disease. Highlights Proposing the combination of medicinal compounds together for the treatment of lung adenocarcinoma achieved a p-value of 0.002858 between lung adenocarcinoma and targeted proteins/genes 3-Leveraging GraphSAGE for Suggesting an Optimal Drug Combinations.