Artificial Intelligence in Medicine: Infrastructures for Deep Learning, Generative Algorithms, and Intelligent Agents

Permanent URI for this collection

https://hdl.handle.net/10125/109917

Browse

Recent Submissions

Now showing 1 - 4 of 4
  • Thumbnail Image
    Item
    Predicting Oncology Readmissions: A Machine Learning Approach Using MIMIC-IV-ED Database
    (2025-01-07) Nascimento, Rayssa; Serra, Cristiano; Nobre De Mello, Ana Carolina; Garcia, Ana Cristina Bicharra
    This study investigates the artificial intelligence (AI) impact on hospital admission classification, specifically focusing on the identification of patients discharged from the emergency room who required readmission within seven days. The research aims to scrutinize potential inconsistencies in human classification within medical triage records and understand the disparities between human decisions and AI predictions. Class balancing techniques like Class Weights and K-Fold Cross-Validation (AI model) and SMOTE, Random Oversampling (The Random Forest) were employed to address imbalanced data, improving the model's ability to generalize across different patient groups. Following the delimitation of cancer cases, the dataset was refined to 7,525 rows, involving 1,945 unique patients. After machine learning training for screening and hospital admission, 80 cases (4.11%) of inconsistent data were observed in triage records, highlighting the critical importance of accurate classification in preventing unnecessary hospital readmissions. The study analyzed key features influencing predictions using the Random Forest algorithm, providing insights to enhance patient management and reduce mortality. The Random Forest classifier exhibited a median accuracy of 72.79% (Mean Precision: 0.76%, Mean Recall: 0.84%, AUC Score: 0.76% and Mean F1-Score: 0.80%), while the neural network showed a Mean Accuracy of 70.3% (Mean Precision: 0.80, Mean Recall: 0.71, AUC Score: 0.75 and Mean F1-Score: 0.75), indicating consistent performance. Feature importance analysis emphasized vital signs, showcasing AI's capability to identify inconsistencies and the need for evolution in triage practices.
  • Thumbnail Image
    Item
    OCC-PAD-OCEAN:An Quantitative Perceptible Modeling of Big Five Personality Based on Computational Affection
    (2025-01-07) Liu, Feng; Hu, Jingyi; Zheng, Qijian
    In the field of human-centred artificial intelligence, there is an increasing focus on the development of AI systems that can be interpreted and quantified from a psychological perspective. This study represents an interdisciplinary fusion of computer science and psychology, with the objective of revolutionising the analysis of personality traits through the application of deep learning techniques. Our research is based on the 'OCC-PAD-OCEAN' approach and employs the robust VGG19 deep learning architecture for the analysis of video data. The error rate of the AI-generated personality prediction is approximately 20% in comparison to the results obtained from the Big Five personality questionnaire. The empirical findings indicate that the predicted values of C (P < 0.001), E (P = 0.005), and A (P = 0.029) dimensions exhibit a significant difference from the measured values (p > 0.05), thereby demonstrating the model's capacity to accurately reflect subtle individual differences within the Big Five traits. It is noteworthy that our analysis revealed minimal gender-related variations (p = 0.611, p = 0.828, p = 0.522, p = 0.696, p = 0.806), yet notable age-related distinctions in traits such as Agreeableness and Neuroticism (p = 0.027 and p = 0.025). The 'OCC-PAD-OCEAN' approach not only overcomes the inherent limitations of traditional questionnaires by providing a more accurate and computationally efficient alternative for psychological evaluations, but also demonstrates the transformative potential of integrating deep learning into psychological analyses.
  • Thumbnail Image
    Item
    Introduction to the Minitrack on Artificial Intelligence in Medicine: Infrastructures for Deep Learning, Generative Algorithms, and Intelligent Agents
    (2025-01-07) Mazurek, Cezary; Polczynska-Bletsos, Monika; Buchwald, Mikolaj; Paderno, Alberto
  • Thumbnail Image
    Item
    A Mixture-of-Experts Decision Support System for Digital Pathology
    (2025-01-07) Kwak, Jin Tae; Bui, Doanh C
    Whole slide image classification is a core task in digital pathology that can assist decision-making procedures for pathologists. Several models, mainly built based upon multiple-instance learning, have shown to be effective in processing and analyzing WSIs. However, these are designed, trained, and evaluated on a single classification task, and thus the models are limited to a specific task and cannot utilize the data and knowledge from other tasks. This substantially limits the ability and expandability of the model to support clinical decision-making. In this study, we present a mixture-of-experts decision support system for digital pathology. The proposed decision support system merges multiple individual models, of which each is tailored to a specific task, and forms a unified model equipped with group intelligence that can handle multiple classification tasks. The proposed system utilizes Transformer architecture to process WSIs and a language decoder to enable flexible classification across multiple tasks. The experimental results on five datasets, including CAMELYON16, TCGA-BRCA, TCGA-NSCLC, TCGA-RCC, and TCGA-ESCA, demonstrate the effectiveness of the proposed approach in supporting clinical decision-making.