ONCOLOGIE, Journal Year: 2024, Volume and Issue: unknown
Published: Dec. 29, 2024
Language: Английский
Breast Cancer Research, Journal Year: 2024, Volume and Issue: 26(1)
Published: Sept. 20, 2024
Language: Английский
Citations
8
Published: Feb. 23, 2024
The smart performance comparison of AI-based breast cancer detection models is an important research topic in the healthcare industry. It used to compare and evaluate different that are diagnose cancer. These mainly developed using machine learning, computer vision, or deep learning techniques. methods these can vary depending on purpose comparison. This include comparing accuracy, precision, recall, f-measure models. Furthermore, other criteria such as stability, reliability, explain ability, speed, cost-effectiveness may be taken into consideration when evaluating have achieved high sensitivity specificity rates, outperforming traditional methods. However, AI varies based type imaging technique dataset used. Further, also highlights need for more diverse inclusive datasets avoid potential biases results from this provide valuable insight help professionals researchers select deploy best model their particular applications.
Language: Английский
Citations
4
Frontiers in Medicine, Journal Year: 2024, Volume and Issue: 11
Published: May 2, 2024
Emerging digital technologies promise to improve breast cancer care, however lack of awareness among clinicians often prevents timely adoption. This study aims investigate current and intention-to-use three healthcare professionals (HCP): (1) health applications (DHA), (2) artificial intelligence (AI), (3) blockchain technology (BC). A 22-item questionnaire was designed administered before after a 30 min educational presentation highlighting implementation examples. Technology were measured using 7-point Likert scales. Correlations between demographics, awareness, intention-to-use, eHealth literacy (GR-eHEALS scale) analyzed. 45 HCP completed the questionnaire, whom 26 (57.8%) female. Age ranged from 24 67 {mean age (SD): 44.93 ± 12.62}. Awareness highest for DHA (68.9%) followed by AI (66.7%) BC (24.4%). The led non-significant increase {5.37 (±1.81) 5.83 (±1.64)}. HCPs´ increased significantly {4.30 (±2.04) 5.90 (±1.67), p < 0.01}. Mean accumulated score GR-eHEALS averaged 33.04 (± 6.61). intended use correlated with (ρ = 0.383; 0.01), 0.591; 0.01) participants´ −0.438; 0.01). demonstrates effect that even short practical can have on emerging technologies. Training potential professional users should be addressed alongside development new information is crucial corresponding use.
Language: Английский
Citations
4
Journal of Cancer Research and Clinical Oncology, Journal Year: 2024, Volume and Issue: 150(8)
Published: Aug. 6, 2024
Abstract Triple negative breast cancer (TNBC) is most aggressive type of with multiple invasive sub-types and leading cause women’s death worldwide. Lack estrogen receptor (ER), progesterone (PR), human epidermal growth factor 2 (HER-2) causes it to spread rapidly making its treatment challenging due unresponsiveness towards anti-HER endocrine therapy. Hence, needing advanced therapeutic treatments strategies in order get better recovery from TNBC. Artificial intelligence (AI) has been emerged by giving high inputs the automated diagnosis as well several diseases, particularly AI based TNBC molecular sub-typing, become successful now days. Therefore, present review reviewed recent advancements role assistance focusing on Meanwhile, advantages, certain limitations future implications are also discussed fully understand readers regarding this issue. Graphical
Language: Английский
Citations
4
Therapeutic Advances in Urology, Journal Year: 2025, Volume and Issue: 17
Published: May 1, 2025
Background: Traditional pathological diagnosis methods have limitations in terms of interobserver variability and the time consumption evaluations. In this study, we explored feasibility using whole-slide images (WSIs) to establish a deep learning model for clear cell renal carcinoma (ccRCC). Methods: We retrospectively collected data from 95 patients with ccRCC January 2023 December 2023. All slices conforming standards were manually annotated first. The WSIs preprocessed extract region interest. divided into training set test set, ratio tumor normal tissue was 3:1. Positive negative samples randomly extracted. Model based on convolutional neural network (CNN) random forest model. accuracy evaluated by generating receiver operating characteristic (ROC) curve. Results: A total 663 collected. mean number per patient 7.6 ± 2.7 (range: 3–17), 506 157 slices. There 200 74 200,870 small 250 63 39,211 According CNN trained 11 identified as false slices, six 94.6% (296/313), precision rate 97.6% (239/245), recall 95.6% (239/250). generated probabilistic heatmaps consistent images. ROC curve results revealed that area under (AUC) reached 0.9658 (95% confidence interval: 0.9603–0.9713), specificity 90.5%, sensitivity 95.6%. Conclusion: use method is feasible. established study achieved high accuracy. AI-based diagnostic may improve efficiency.
Language: Английский
Citations
0
Journal of Education and Educational Research, Journal Year: 2024, Volume and Issue: 11(3), P. 34 - 38
Published: Dec. 10, 2024
Oxymatrine (OMT), an alkaloid derived from Sophora flavescens, possesses a diverse array of pharmacological activities, encompassing anti-inflammatory, antiviral, antitumor, and immune regulatory properties. In recent times, significant strides have been made in investigating the efficacy OMT against breast cancer. exerts its anticancer influence through various mechanisms, including inhibition tumor cell proliferation, suppression invasion migration, induction apoptosis, disruption cycle. This article systematically reviews underlying mechanisms cancer therapy by conducting comprehensive search across multiple databases, such as CNKI, Wanfang, PubMed, utilizing keywords like "OMT," "breast cancer," "pharmacology," "pharmacokinetics," well incorporating both Chinese English terms for "triple-negative cancer." The objective is to delve into potential applications treatment furnish valuable reference future clinical investigations exploring utilization this field.
Language: Английский
Citations
0
Published: Jan. 11, 2024
The aim of the project is to compare performance four different machine learning algorithms for breast cancer prediction such as decision tree, logistic regression, XG boost, and CAT boost. We used a dataset patient medical records containing various clinical factors train test algorithms. Accurate diagnosis early detection are essential enhancing outcomes. Data mining prominent tool in healthcare industry processing massive amounts data. To examine complicated data, researchers use variety data approaches. these strategies can help practitioners forecast onset cancer.
Language: Английский
Citations
0
Published: May 29, 2024
The purpose of this study was to investigate whether the lesion risk score provided by an AI system is influenced selection exposure parameters. A breast phantom which contains a lesion, imaged with digital mammography different imaging conditions. tube voltage, dose level and anode-filter combination were varied based on obtained automatic control. organ for each image extracted from DICOM header. images analyzed system, (suspicion malignancy) condition. Correlations between conditions investigated. results showed that had strong impact score. Reducing low resulted in no longer detected lesion. Images suboptimal quality may result inaccurate performance. In our preliminary analysis, proven be realistic enough being system.
Language: Английский
Citations
0
IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 134635 - 134650
Published: Jan. 1, 2024
Breast cancer poses a global health challenge, with high incidence and mortality rates. Early detection precise diagnosis are crucial for patient prognosis. Machine learning (ML) models applied to mammary biopsy image data hold promise achieving an efficient accurate breast diagnosis. In this study, we evaluated the performance of several ML algorithms, including Logistic Regression (LR), Random Forest (RF), Naive Bayes (NB) Support Vector (SVM). We establish evaluation contexts by implementing standardization reducing correlation between variables. Firstly, select best-performing parameters each algorithm building evaluating individual models. Then, implement combined model using weighted voting, where weights determined based on its test dataset. The final is constructed combining LR, RF SVM find that best-performance model, so it has highest weight in model. integrated achieves accuracy 98%, precision 97%, recall 99%, F1-score 98% AUC 0.98. Our voting compares favourably other analysed. This approach demonstrates efficiency transparency handling structured medical data. It prototype will be refined expanded encompass larger real-world datasets.
Language: Английский
Citations
0
Cancer Informatics, Journal Year: 2024, Volume and Issue: 23
Published: Jan. 1, 2024
This narrative review explores the burgeoning field of Artificial Intelligence (AI)-driven Breast Cancer (BC) survival prediction, emphasizing transformative impact on patient care. From machine learning to deep neural networks, diverse models demonstrate potential refine prognosis accuracy and tailor treatment strategies. The literature underscores need for clinician integration addresses challenges model generalizability ethical considerations. Crucially, AI's promise extends Low- Middle-Income Countries (LMICs), presenting an opportunity bridge healthcare disparities. Collaborative efforts in research, technology transfer, education are essential empower professionals LMICs. As we navigate this frontier, AI emerges not only as a technological advancement but guiding light toward personalized, accessible BC care, marking significant stride global fight against formidable disease.
Language: Английский
Citations
0