Small molecule-mediated targeting of microRNAs for drug discovery: Experiments, computational techniques, and disease implications DOI Creative Commons
Jianfeng Sun,

Miaoer Xu,

Jinlong Ru

et al.

European Journal of Medicinal Chemistry, Journal Year: 2023, Volume and Issue: 257, P. 115500 - 115500

Published: May 17, 2023

Small molecules have been providing medical breakthroughs for human diseases more than a century. Recently, identifying small molecule inhibitors that target microRNAs (miRNAs) has gained importance, despite the challenges posed by labour-intensive screening experiments and significant efforts required medicinal chemistry optimization. Numerous experimentally-verified cases demonstrated potential of miRNA-targeted disease treatment. This new approach is grounded in their posttranscriptional regulation expression disease-associated genes. Reversing dysregulated gene using this mechanism may help control dysfunctional pathways. Furthermore, ongoing improvement algorithms allowed integration computational strategies built on top laboratory-based data, facilitating precise rational design discovery lead compounds. To complement use extensive pharmacogenomics data prioritising drugs, our previous work introduced based only molecular sequences. Moreover, various tools predicting interactions biological networks similarity-based inference techniques accumulated established studies. However, there are limited number comprehensive reviews covering both experimental drug processes. In review, we outline cohesive overview applications discovery, along with implications clinical significance. Finally, utilizing drug-target interaction (DTIs) from DrugBank, showcase effectiveness deep learning obtaining physicochemical characterization DTIs.

Language: Английский

A survey on deep learning applied to medical images: from simple artificial neural networks to generative models DOI Open Access
Pedro Celard, Eva Iglesias, José Manuel Sorribes-Fdez

et al.

Neural Computing and Applications, Journal Year: 2022, Volume and Issue: 35(3), P. 2291 - 2323

Published: Nov. 4, 2022

Language: Английский

Citations

101
Breast cancer diagnosis from histopathology images using deep neural network and XGBoost DOI
Alireza Maleki, Mohammad Raahemi, Hamid Nasiri

et al.

Biomedical Signal Processing and Control, Journal Year: 2023, Volume and Issue: 86, P. 105152 - 105152

Published: June 19, 2023

Language: Английский

Citations

51
Small and overlapping worker detection at construction sites DOI Creative Commons
Minsoo Park, Dai Quoc Tran, JinYeong Bak

et al.

Automation in Construction, Journal Year: 2023, Volume and Issue: 151, P. 104856 - 104856

Published: April 12, 2023

Although there has been study on worker detection using computer vision (CV) for the safety of construction sites, it is still challenging to identify employees who are obstructed or have poor vision. To solve these problems, we propose a method small and overlapping target (worker) at complex site named SOC-YOLO. The based YOLOv5 utilizes distance intersection over union (DIoU) non-maximum suppression (NMS), incorporating weighted triplet attention, expansion feature-level, Soft-pool. Workers can be captured with overlap, particularly in large-scale DIoU-based loss function, NMS contributed accuracy improvement. Next, weighted-triplet attention mechanism that extract feature information from space more effectively channel when learning object networks, simple average approach same weight between existing attention. model adds additional predictive heads residual connections address workers photographed long distances. A low-level map containing regarding targets used by extending level. Finally, Softpool-spatial pyramid pooling fast (Softpool-SPPF) proposed problem inconsistent input image sizes. Softpool-SPPF performs an spatial (SPP) function while preserving functional accurate detection. Experiments were conducted published datasets handmade datasets, results showed increase 81.26% 84.63% precision (AP) objects, 67.52% 73.88% mAP minute 74.56% to77.57% objects. expected useful monitoring applying tracking model.

Language: Английский

Citations

44
Alzheimer’s Disease Detection Using Deep Learning on Neuroimaging: A Systematic Review DOI Creative Commons
Mohammed Alsubaie, Suhuai Luo, Kamran Shaukat

et al.

Machine Learning and Knowledge Extraction, Journal Year: 2024, Volume and Issue: 6(1), P. 464 - 505

Published: Feb. 21, 2024

Alzheimer’s disease (AD) is a pressing global issue, demanding effective diagnostic approaches. This systematic review surveys the recent literature (2018 onwards) to illuminate current landscape of AD detection via deep learning. Focusing on neuroimaging, this study explores single- and multi-modality investigations, delving into biomarkers, features, preprocessing techniques. Various models, including convolutional neural networks (CNNs), recurrent (RNNs), generative are evaluated for their performance. Challenges such as limited datasets training procedures persist. Emphasis placed need differentiate from similar brain patterns, necessitating discriminative feature representations. highlights learning’s potential limitations in detection, underscoring dataset importance. Future directions involve benchmark platform development streamlined comparisons. In conclusion, while learning holds promise accurate refining models methods crucial tackle challenges enhance precision.

Language: Английский

Citations

27
LeafConvNeXt: Enhancing plant disease classification for the future of unmanned farming DOI

Feifei Lu,

Hong Shangguan,

Yizhe Yuan

et al.

Computers and Electronics in Agriculture, Journal Year: 2025, Volume and Issue: 233, P. 110165 - 110165

Published: March 6, 2025

Language: Английский

Citations

2
An attention-guided convolutional neural network for automated classification of brain tumor from MRI DOI
Sumeet Saurav, Ayush Sharma, Ravi Saini

et al.

Neural Computing and Applications, Journal Year: 2022, Volume and Issue: 35(3), P. 2541 - 2560

Published: Sept. 1, 2022

Language: Английский

Citations

41
AD-BERT: Using pre-trained language model to predict the progression from mild cognitive impairment to Alzheimer's disease DOI Creative Commons
Chengsheng Mao, Jie Xu, Luke V. Rasmussen

et al.

Journal of Biomedical Informatics, Journal Year: 2023, Volume and Issue: 144, P. 104442 - 104442

Published: July 8, 2023

Language: Английский

Citations

25
A novel image expression-driven modeling strategy for coke quality prediction in the smart cokemaking process DOI Creative Commons
Yuhang Qiu,

Yunze Hui,

Pengxiang Zhao

et al.

Energy, Journal Year: 2024, Volume and Issue: 294, P. 130866 - 130866

Published: March 7, 2024

In pursuit of carbon neutrality and advancing energy-efficient practices within the steel coking industries, traditional cokemaking process is progressively evolving towards intelligence, with coke quality prediction emerging as a pivotal technology at its core. Nevertheless, intricacy production presents formidable challenge in accurately forecasting it. This study first to propose novel image expression-driven modeling approach that transforms numerical coal properties into expressions uniquely integrates utilization convolutional neural network (CNN) for predicting including strength after reaction (CSR) reactivity index (CRI). Utilizing collected 729 Chinese corresponding indexes, dimensionality reduction technique was employed transform expressions. A combined random forest model subsequently developed learning prediction, performance evaluated on root mean squared error (RMSE), absolute (MAE), R2 metrics. The results suggested proposed groundbreaking outperformed existing properties-based models typical regression models, achieving MAE 1.57, RMSE 2.22, 0.86 metric, along 1.82 2.42 well 0.91 metric CRI CSR respectively. Furthermore, comprehensive analysis also undertaken identify factors influencing efficacy based approach.

Language: Английский

Citations

15
Shallow and deep learning classifiers in medical image analysis DOI Creative Commons
Francesco Prinzi, Tiziana Currieri, Salvatore Gaglio

et al.

European Radiology Experimental, Journal Year: 2024, Volume and Issue: 8(1)

Published: March 5, 2024

Abstract An increasingly strong connection between artificial intelligence and medicine has enabled the development of predictive models capable supporting physicians’ decision-making. Artificial encompasses much more than machine learning, which nevertheless is its most cited used sub-branch in last decade. Since clinical problems can be modeled through learning classifiers, it essential to discuss their main elements. This review aims give primary educational insights on accessible widely employed classifiers radiology field, distinguishing “shallow” ( i.e., traditional learning) algorithms, including support vector machines, random forest XGBoost, “deep” architectures convolutional neural networks vision transformers. In addition, paper outlines key steps for training highlights differences common algorithms architectures. Although choice an algorithm depends task dataset dealing with, general guidelines classifier selection are proposed relation analysis, size, explainability requirements, available computing resources. Considering enormous interest these innovative architectures, problem interpretability finally discussed, providing a future perspective trustworthy intelligence. Relevance statement The growing synergy fosters aiding physicians. Machine from shallow deep offering crucial decision systems healthcare. Explainability feature that leads toward integration into practice. Key points • Training requires extracting disease-related features region interests e.g., radiomics). Deep implement automatic extraction classification. based data computational resources availability, task, explanation needs. Graphical

Language: Английский

Citations

11
A Hybrid Learning-Architecture for Improved Brain Tumor Recognition DOI Creative Commons
J. R. Dixon, Oluwatunmise Akinniyi, Abeer Abdelhamid

et al.

Algorithms, Journal Year: 2024, Volume and Issue: 17(6), P. 221 - 221

Published: May 21, 2024

The accurate classification of brain tumors is an important step for early intervention. Artificial intelligence (AI)-based diagnostic systems have been utilized in recent years to help automate the process and provide more objective faster diagnosis. This work introduces enhanced AI-based architecture improved tumor classification. We introduce a hybrid that integrates vision transformer (ViT) deep neural networks create ensemble classifier, resulting robust framework. analysis pipeline begins with preprocessing data normalization, followed by extracting three types MRI-derived information-rich features. latter included higher-order texture structural feature sets harness spatial interactions between image intensities, which were derived using Haralick features local binary patterns. Additionally, deeper images are extracted optimized convolutional (CNN) architecture. Finally, ViT-derived also integrated due their ability handle dependencies across larger distances while being less sensitive augmentation. then weighted, fused, fed machine learning classifier final MRIs. proposed weighted has evaluated on publicly available locally collected MRIs four classes various metrics. results showed leveraging benefits individual components leads performance ablation studies.

Language: Английский

Citations

10