Computer Vision to Enhance Healthcare Domain: An Overview of Features, Implementation, and Opportunities

Intelligent Pharmacy, Год журнала: 2024, Номер 2(6), С. 792 - 803

Опубликована: Май 21, 2024

The emergence of Artificial Intelligence (AI) has already brought several advantages to the healthcare sector. Computer Vision (CV) is one growing modern AI technologies. distribution and administration medications are about change by using CV for medication management. This system scans pharmaceutical labels keeps track process from delivery cameras, sensors, computer algorithms. In order assure accuracy in medicine dose, also makes it easier doctors, nurses, chemists communicate. vision-driven management can significantly lower number medical mistakes that result inaccurate or missing prescriptions, improper doses, simply forgetting take a particular drug. An exhaustive literature review been done identify work related research objectives. paper their need healthcare. Various tasks associated with domain discussed. Targeted goals through traits briefed. Finally, significant applications CVs were identified Nowadays, practical uses Its methods widely used since they have shown excellent utility contexts, including imaging surgical planning. study how program computers comprehend digital pictures. Numerous utilise this technology, such as automated abnormality identification, illness diagnosis, procedure guiding. expanding quickly enormous promise enhance Some many sector include patient identification systems, picture analysis, simulation diagnosis.

Язык: Английский

---

Artificial intelligence‐based motion tracking in cancer radiotherapy: A review

Journal of Applied Clinical Medical Physics, Год журнала: 2024, Номер 25(11)

Опубликована: Авг. 28, 2024

Abstract Radiotherapy aims to deliver a prescribed dose the tumor while sparing neighboring organs at risk (OARs). Increasingly complex treatment techniques such as volumetric modulated arc therapy (VMAT), stereotactic radiosurgery (SRS), body radiotherapy (SBRT), and proton have been developed doses more precisely target. While technologies improved delivery, implementation of intra‐fraction motion management verify position time has become increasingly relevant. Artificial intelligence (AI) recently demonstrated great potential for real‐time tracking tumors during treatment. However, AI‐based faces several challenges, including bias in training data, poor transparency, difficult data collection, workflows quality assurance, limited sample sizes. This review presents AI algorithms used chest, abdomen, pelvic management/tracking provides literature summary on topic. We will also discuss limitations these studies propose improvements.

Язык: Английский

---

Precision Medicine for Apical Lesions and Peri-Endo Combined Lesions Based on Transfer Learning Using Periapical Radiographs

Bioengineering, Год журнала: 2024, Номер 11(9), С. 877 - 877

Опубликована: Авг. 29, 2024

An apical lesion is caused by bacteria invading the tooth apex through caries. Periodontal disease plaque accumulation. Peri-endo combined lesions include both diseases and significantly affect dental prognosis. The lack of clear symptoms in early stages onset makes diagnosis challenging, delayed treatment can lead to spread symptoms. Early infection detection crucial for preventing complications. PAs used as database were provided Chang Gung Memorial Medical Center, Taoyuan, Taiwan, with permission from Institutional Review Board (IRB): 02002030B0. image enhancement method a new technology PA detection. This convolutional neural networks (CNN) classify lesions, peri-endo asymptomatic cases, compare You Only Look Once-v8-Oriented Bounding Box (YOLOv8-OBB) results. contributions lie utilization augmentation adaptive histogram equalization on individual images, achieving highest comprehensive validation accuracy 95.23% ConvNextv2 model. Furthermore, CNN outperformed YOLOv8 identifying an F1-Score 92.45%. For classification attained 96.49%, whereas scored 88.49%.

Язык: Английский

---

The Diagnostic Classification of the Pathological Image Using Computer Vision

Algorithms, Год журнала: 2025, Номер 18(2), С. 96 - 96

Опубликована: Фев. 8, 2025

Computer vision and artificial intelligence have revolutionized the field of pathological image analysis, enabling faster more accurate diagnostic classification. Deep learning architectures like convolutional neural networks (CNNs), shown superior performance in tasks such as classification, segmentation, object detection pathology. has significantly improved accuracy disease diagnosis healthcare. By leveraging advanced algorithms machine techniques, computer systems can analyze medical images with high precision, often matching or even surpassing human expert performance. In pathology, deep models been trained on large datasets annotated pathology to perform cancer diagnosis, grading, prognostication. While approaches show great promise challenges remain, including issues related model interpretability, reliability, generalization across diverse patient populations imaging settings.

Язык: Английский

---

Advanced imaging techniques and artificial intelligence in pleural diseases: a narrative review

European Respiratory Review, Год журнала: 2025, Номер 34(176), С. 240263 - 240263

Опубликована: Апрель 1, 2025

Background Pleural diseases represent a significant healthcare burden, affecting over 350 000 patients annually in the US alone and requiring accurate diagnostic approaches for optimal management. Traditional imaging techniques have limitations differentiating various pleural disorders invasive procedures are usually required definitive diagnosis. Methods We conducted nonsystematic, narrative literature review aimed at describing latest advances artificial intelligence (AI) applications diseases. Results Novel ultrasound-based techniques, such as elastography contrast-enhanced ultrasound, described their promising accuracy malignant from benign lesions. Quantitative utilising pixel-density measurements to noninvasively distinguish exudative transudative effusions highlighted. AI algorithms, which shown remarkable performance abnormality detection, effusion characterisation automated fluid volume quantification, also described. Finally, role of deep-learning models early complication detection analysis follow-up studies is examined. Conclusions Advanced show promise management diseases, improving reducing need procedures. However, larger prospective needed validation. The integration AI-driven with molecular genomic data offers potential personalised therapeutic strategies, although challenges privacy, algorithm transparency clinical validation persist. This comprehensive approach may revolutionise disease management, enhancing patient outcomes through more accurate, noninvasive strategies.

Язык: Английский

---

Automated engineered-stone silicosis screening and staging using Deep Learning with X-rays

Computers in Biology and Medicine, Год журнала: 2025, Номер 191, С. 110153 - 110153

Опубликована: Апрель 18, 2025

Silicosis, a debilitating occupational lung disease caused by inhaling crystalline silica, continues to be significant global health issue, especially with the increasing use of engineered stone (ES) surfaces containing high silica content. Traditional diagnostic methods, dependent on radiological interpretation, have low sensitivity, especially, in early stages disease, and present variability between evaluators. This study explores efficacy deep learning techniques automating screening staging silicosis using chest X-ray images. Utilizing comprehensive dataset, obtained from medical records cohort workers exposed artificial quartz conglomerates, we implemented preprocessing stage for rib-cage segmentation, followed classification state-of-the-art models. The segmentation model exhibited precision, ensuring accurate identification thoracic structures. In phase, our models achieved near-perfect accuracy, ROC AUC values reaching 1.0, effectively distinguishing healthy individuals those silicosis. demonstrated remarkable precision disease. Nevertheless, differentiating simple progressive massive fibrosis, evolved complicated form presented certain difficulties, during transitional period, when assessment can significantly subjective. Notwithstanding these an accuracy around 81% scores nearing 0.93. highlights potential generate clinical decision support tools increase effectiveness diagnosis silicosis, whose detection would allow patient moved away all sources exposure, therefore constituting substantial advancement diagnostics.

Язык: Английский

---

Efficient YOLO Based Deep Learning Model for Arabic Sign Language Recognition

Research Square (Research Square), Год журнала: 2024, Номер unknown

Опубликована: Март 28, 2024

Verbal communication is the dominant form of self-expression and interpersonal communication. Speech a considerable obstacle for individuals with disabilities, including those who are deaf, hard hearing, mute, or nonverbal. Consequently, these depend on sign language to communicate others. Sign Language complex system gestures visual cues that facilitate inclusion into vocal groups. In this manuscript novel technique proposed using deep learning recognize Arabic (ArSL) accurately. Through advanced system, objective help in between hearing deaf community. The mechanism relies attention mechanisms, state-of-art Convolutional Neural Network (CNN) architectures robust YOLO object detection model highly improves implementation accuracy ArSL recognition. our method, we integrate self-attention block, channel module, spatial cross-convolution module features processing, recognition reaches 98.9%. method significantly improved higher rate. presented approach showed significant improvement as compared conventional techniques precision rate 0.9. For [email protected], mAP score 0.9909 while [email protected]:0.95 results tops all state-of-the-art techniques. This shows has great capability accurately detect classify multiple signs. provides unique way linking people improving strategy also promoting social region.

Язык: Английский

---

Efficient YOLO-Based Deep Learning Model for Arabic Sign Language Recognition

Deleted Journal, Год журнала: 2024, Номер 3(4)

Опубликована: Май 7, 2024

Verbal communication is the dominant form of self-expression and interpersonal communication. Speech a considerable obstacle for individuals with disabilities, including those who are deaf, hard hearing, mute, nonverbal. Sign language complex system gestures visual signs facilitating individual With help artificial intelligence, hearing deaf can communicate more easily. Automatic detection recognition sign challenging task in computer vision machine learning. This paper proposes novel technique using deep learning to recognize Arabic Language (ArSL) accurately. The proposed method relies on advanced attention mechanisms convolutional neural network architecture integrated robust You Only Look Once (YOLO) object model that improves rate technique. In our method, we integrate self-attention block, channel module, spatial cross-convolution module into feature processing accurate detection. accuracy significantly improved, higher 99%. methodology outperformed conventional methods, achieving precision 0.9 mean average (mAP) 0.9909 at an intersection over union (IoU) 0.5. From IoU thresholds 0.5 0.95, mAP continuously remains high, indicating its effectiveness accurately identifying different levels. results show model’s robustness detecting classifying multiple ArSL signs. efficacy model.

Язык: Английский

---

Enhanced Radiological Anomaly Detection using Optimized YOLO-NAS Model

Опубликована: Май 16, 2024

Язык: Английский

---

Enhancing Pulmonary Diagnosis in Chest X-rays through Generative AI Techniques

J — Multidisciplinary Scientific Journal, Год журнала: 2024, Номер 7(3), С. 302 - 318

Опубликована: Авг. 13, 2024

Chest X-ray imaging is an essential tool in the diagnostic procedure for pulmonary conditions, providing healthcare professionals with capability to immediately and accurately determine lung anomalies. This modality fundamental assessing confirming presence of various issues, allowing timely effective medical intervention. In response widespread prevalence infections globally, there a growing imperative adopt automated systems that leverage deep learning (DL) algorithms. These are particularly adept at handling large radiological datasets high precision. study introduces advanced identification model utilizes VGG16 architecture, specifically adapted identifying anomalies such as opacity, COVID-19 pneumonia, normal appearance lungs, viral pneumonia. Furthermore, we address issue generalizability, which prime significance our work. We employed data augmentation technique through CycleGAN, which, experimental outcomes, has proven enhancing robustness model. The combined performance VGG CycleGAN demonstrates remarkable outcomes several evaluation metrics, including recall, F1-score, accuracy, precision, area under curve (AUC). results showcased achieving 98.58%. contributes advancing generative artificial intelligence (AI) analysis establishes solid foundation ongoing developments computer vision technologies within sector.

Язык: Английский

---