METHODS: This quasi-experimental quantitative study was conducted among Year 1 MBBS students at Universiti Kuala Lumpur Royal College of Medicine Perak, Malaysia. Freely available AI tools, including ChatGPT accessed through the free web-based interface and other student-selected AI applications such as AI summarizers, quiz makers, flashcard generators, and slide assistants, were introduced as optional adjuncts during one two-hour end-of-module respiratory pharmacology tutorial after completion of the six-week respiratory module. Pre-test and post-test multiple-choice questions were used to summarize knowledge scores descriptively. A content-validated structured questionnaire comprising five thematic domains on a five-point Likert scale was used to assess student perceptions. Descriptive statistics were performed using means, standard deviations, frequencies and percentages. For Likert-scale items, mean scores and standard deviations were calculated for each item and the grouped domain.

RESULTS: Sixty-nine Year 1 MBBS students completed the perception survey. The Moodle view count was not used as the response-rate denominator because it reflected page/link views rather than verified eligible consenting participants. Pre-test and post-test mean scores were 9.61/10 and 9.52/10, respectively, indicating no meaningful descriptive change in knowledge. Overall perceptions of AI-assisted learning were positive (mean 4.15/5). The highest-rated benefits were improved understanding of pharmacology topics (mean 4.43), time-saving during self-study (mean 4.36), more effective doubt clarification (mean 4.31), and willingness to recommend AI tools for other subjects (mean 4.30). Students most frequently used AI to explain difficult concepts and to summarize lecture content. Though ChatGPT was rated the most helpful tool, there was concern about incorrect or outdated information. A total of 63.5% of respondents stated that they would continue using AI because it saves time and improves study efficiency.

METHODS: In this paper, we propose a novel approach to distinguish colonic polyps by integrating several techniques, including a modified deep residual network, principal component analysis and AdaBoost ensemble learning. A powerful deep residual network architecture, ResNet-50, was investigated to reduce the computational time by altering its architecture. To keep the interference to a minimum, median filter, image thresholding, contrast enhancement, and normalisation techniques were exploited on the endoscopic images to train the classification model. Three publicly available datasets, i.e., Kvasir, ETIS-LaribPolypDB, and CVC-ClinicDB, were merged to train the model, which included images with and without polyps.

RESULTS: The proposed approach trained with a combination of three datasets achieved Matthews Correlation Coefficient (MCC) of 0.9819 with accuracy, sensitivity, precision, and specificity of 99.10%, 98.82%, 99.37%, and 99.38%, respectively.

METHODS: We investigated the existing body of evidence and applied Preferred Reporting Items for Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to search records in IEEE, Google scholar, and PubMed databases. We identified 65 papers that were published from 2013 to 2022 and these papers cover 67 different studies. The review process was structured according to the medical data that was used for disease detection. We identified six main categories, namely air flow, genetic, imaging, signals, and miscellaneous. For each of these categories, we report both disease detection methods and their performance.

RESULTS: We found that medical imaging was used in 14 of the reviewed studies as data for automated obstructive airway disease detection. Genetics and physiological signals were used in 13 studies. Medical records and air flow were used in 9 and 7 studies, respectively. Most papers were published in 2020 and we found three times more work on Machine Learning (ML) when compared to Deep Learning (DL). Statistical analysis shows that DL techniques achieve higher Accuracy (ACC) when compared to ML. Convolutional Neural Network (CNN) is the most common DL classifier and Support Vector Machine (SVM) is the most widely used ML classifier. During our review, we discovered only two publicly available asthma and COPD datasets. Most studies used private clinical datasets, so data size and data composition are inconsistent.