A systematic review of trustworthy and explainable artificial intelligence in healthcare: Assessment of quality, bias risk, and data fusion

Information Fusion, Journal Year: 2023, Volume and Issue: 96, P. 156 - 191

Published: March 15, 2023

Language: Английский

---

Explainable AI-driven IoMT fusion: Unravelling techniques, opportunities, and challenges with Explainable AI in healthcare

Information Fusion, Journal Year: 2024, Volume and Issue: 110, P. 102472 - 102472

Published: May 16, 2024

Language: Английский

---

A comprehensive review of elderly fall detection using wireless communication and artificial intelligence techniques

Measurement, Journal Year: 2024, Volume and Issue: 226, P. 114186 - 114186

Published: Jan. 20, 2024

Language: Английский

---

A Robust and Automated Vision-Based Human Fall Detection System Using 3D Multi-Stream CNNs with an Image Fusion Technique

Applied Sciences, Journal Year: 2023, Volume and Issue: 13(12), P. 6916 - 6916

Published: June 7, 2023

Unintentional human falls, particularly in older adults, can result severe injuries and death, negatively impact quality of life. The World Health Organization (WHO) states that falls are a significant public health issue the primary cause injury-related fatalities worldwide. Injuries resulting from such as broken bones, trauma, internal injuries, have consequences lead to loss mobility independence. To address this problem, there been suggestions develop strategies reduce frequency order decrease healthcare costs productivity loss. Vision-based fall detection approaches proven their effectiveness addressing on time, which help injuries. This paper introduces an automated vision-based system for detecting issuing instant alerts upon detection. proposed processes live footage monitoring surveillance camera by utilizing fine-tuned segmentation model image fusion technique pre-processing classifying set with 3D multi-stream CNN (4S-3DCNN). when sequence Falling monitored human, followed having Fallen, takes place. was assessed using publicly available Le2i dataset. System validation revealed impressive result, achieving accuracy 99.44%, sensitivity 99.12%, specificity precision 99.59%. Based reported results, presented be valuable tool preventing injury complications, reducing costs.

Language: Английский

---

Human Fall Detection Using Transfer Learning-Based 3D CNN

Communications in computer and information science, Journal Year: 2025, Volume and Issue: unknown, P. 102 - 113

Published: Jan. 1, 2025

Language: Английский

---

Integrating IoT and Machine Learning for Human Fall Detection and Activity Monitoring

Communications in computer and information science, Journal Year: 2025, Volume and Issue: unknown, P. 25 - 40

Published: Jan. 1, 2025

Language: Английский

---

A video-based fall detection using 3D sparse convolutional neural network in elderly care services

Machine Graphics and Vision, Journal Year: 2025, Volume and Issue: 34(1), P. 53 - 74

Published: March 28, 2025

Falls in the elderly have become one of major risks for growing population. Therefore, application automatic fall detection system is particularly important. In recent years, a large number deep learning methods (such as CNN) been applied to such research. This paper proposed sparse convolution method 3D Sparse Convolutions and corresponding Convolutional Neural Network (3D-SCNN), which can achieve faster at approximate accuracy, thereby reducing computational complexity while maintaining high accuracy video analysis task. Additionally, preprocessing stage involves dynamic key frame selection method, using jitter buffers adjust based on current network conditions buffer state. To ensure feature continuity, overlapping cubes selected frames are intentionally employed, with resizing adapt dynamics states. Experiments conducted Multi-camera dataset UR dataset, results show that its exceeds three compared methods, outperforms traditional 3D-CNN both losses.

Language: Английский

---

Multimodal Fall Detection Using Spatial–Temporal Attention and Bi-LSTM-Based Feature Fusion

Future Internet, Journal Year: 2025, Volume and Issue: 17(4), P. 173 - 173

Published: April 15, 2025

Human fall detection is a significant healthcare concern, particularly among the elderly, due to its links muscle weakness, cardiovascular issues, and locomotive syndrome. Accurate crucial for timely intervention injury prevention, which has led many researchers work on developing effective systems. However, existing unimodal systems that rely solely skeleton or sensor data face challenges such as poor robustness, computational inefficiency, sensitivity environmental conditions. While some multimodal approaches have been proposed, they often struggle capture long-range dependencies effectively. In order address these challenges, we propose framework integrates data. The system uses Graph-based Spatial-Temporal Convolutional Attention Neural Network (GSTCAN) spatial temporal relationships from motion information in stream-1, while Bi-LSTM with Channel (CA) processes stream-2, extracting both features. GSTCAN model AlphaPose extraction, calculates between consecutive frames, applies graph convolutional network (GCN) CA mechanism focus relevant features suppressing noise. parallel, inertial signals, capturing refining feature representations. branches are fused passed through fully connected layer classification, providing comprehensive understanding of human motion. proposed was evaluated Fall Up UR datasets, achieving classification accuracy 99.09% 99.32%, respectively, surpassing methods. This robust efficient demonstrates strong potential accurate continuous monitoring.

Language: Английский

---

Indoor Monitoring System Based on Computer Vision for Fall Detection Oriented to Elderly Assistance

Lecture notes in networks and systems, Journal Year: 2025, Volume and Issue: unknown, P. 124 - 138

Published: Jan. 1, 2025

Language: Английский

---

Fall Recognition Based on Time-Level Decision Fusion Classification

Applied Sciences, Journal Year: 2024, Volume and Issue: 14(2), P. 709 - 709

Published: Jan. 14, 2024

We propose a vision-based fall detection algorithm using advanced deep learning models and fusion methods for smart safety management systems. By detecting falls through visual cues, it is possible to leverage existing surveillance cameras, thus minimizing the need extensive additional equipment. Consequently, we developed cost-effective system. The proposed system consists of four modules: object detection, pose estimation, action recognition, result fusion. Constructing involved utilization state-of-the-art (SOTA) models. In module, experimented with various approaches, including voting, maximum, averaging, probabilistic Notably, observed significant performance improvement use employed HAR-UP dataset demonstrate this enhancement, achieving an average 0.84% increase in accuracy compared baseline, which did not incorporate methods. applying our time-level ensemble skeleton-based approach, coupled enhanced estimation modules, substantially improved robustness system, particularly challenging scenarios.

Language: Английский

---