A journal dedicated to the full breadth of medical and health science scholarship — from clinical theory and biomedical research to health policy and the transformation of medicine through technology.
Journal of Medical Theory and Technology is published by Airela Press as a peer-reviewed journal. We publish original research, clinical studies, systematic reviews, theoretical essays, and case reports spanning all areas of medicine and the health sciences.
We welcome contributions from clinicians, biomedical scientists, public health researchers, health educators, medical ethicists, and policy scholars. Interdisciplinary submissions that cross the boundaries of medicine, technology, social science, and the humanities are particularly encouraged.
Our scope is broad by design: wherever rigorous scholarship advances understanding of health, disease, care, or the medical sciences, it belongs here.
Philosophy of medicine, medical ethics, history of medicine, theoretical frameworks in clinical science, and the conceptual foundations of health and disease.
Clinical research, patient care, epidemiology, public health interventions, health systems, nursing and allied health, medical education, and health policy.
Medical technology, AI in healthcare, digital health, precision medicine, biomedical engineering, telemedicine, and the ethical dimensions of healthcare innovation.
We invite submissions from clinicians, researchers, and health scholars whose work advances understanding of medicine, health, clinical practice, or biomedical science.
Manuscripts should be submitted by email with the subject line "Submission — [Paper Title]". Please attach your manuscript as a PDF or Word document.
All submissions undergo double-blind peer review. Authors can expect an initial decision within four to six weeks of submission. We accept original research articles, review articles, theoretical essays, and empirical studies.
Airela Press and the Journal of Medical Theory and Technology are committed to the highest standards of publication ethics. We follow the guidelines established by the Committee on Publication Ethics (COPE).
All submissions are screened for plagiarism prior to peer review. Fabrication, falsification, and misrepresentation of data or results will result in immediate rejection and may be reported to the authors' institutions.
Authors are required to disclose any conflicts of interest relevant to their submission. Funding sources should be acknowledged transparently in all published work.
Reviewers are expected to maintain strict confidentiality regarding manuscripts under review and to disclose any conflicts of interest to the editors before undertaking review.
Corrections and retractions will be issued promptly and transparently whenever errors are identified in published work. The journal maintains a permanent public record of all corrections.
Authors retain copyright to their published work under a Creative Commons Attribution (CC BY 4.0) licence, which permits free reuse with appropriate attribution.
This study proposes a novel method based on deep learning and multi-feature fusion for the recognition of herbaceous Tibetan medicine slice images in complex backgrounds. A dataset comprising 3,200 images of 32 types of Tibetan medicines was constructed and expanded to 22,400 images through data augmentation techniques. For feature extraction, the method integrates RGB color features, improved HOG (GHOG) shape features, and LBP texture features. Deep learning was conducted using an improved AlexNet model (M-AlexNet) with an attention mechanism, and a multi-feature fusion strategy was employed to optimize recognition performance. Experimental results demonstrate that the proposed method achieves a recognition accuracy of 92.31% in complex backgrounds, significantly outperforming other comparative methods. The introduction of data augmentation and GHOG features improved recognition accuracy by 1.94% and 3.42%, respectively. This study provides an effective solution for the automatic recognition of Tibetan medicine slices in complex environments, holding significant implications for research and applications in Tibetan medicine.
The rapid development of AI-assisted diagnosis is profoundly transforming both the epistemic structure of medical diagnosis and the allocation of responsibility within clinical practice. Unlike traditional medical tools, machine-learning–based diagnostic systems engage deeply in clinical reasoning through algorithmic opacity, continuous learning, and probabilistic outputs, thereby challenging established models of medical responsibility centered on human physicians. When errors occur in AI-assisted diagnosis and result in patient harm, the question of how responsibility should be attributed has become a pressing issue in medical ethics and health law. From a normative ethical perspective, this article systematically examines the distinctive challenges of responsibility attribution in AI-assisted diagnosis. It first elucidates the technical features that position diagnostic AI systems as “cognitive collaborators,” blurring the traditional boundary between instruments and decision-making agents. The article then comparatively analyzes major theoretical approaches to responsibility attribution, including physician-centered liability, manufacturer responsibility, distributed responsibility, and no-fault compensation schemes, demonstrating that no single model adequately captures the complex socio-technical realities involved. Through illustrative case analysis, the article further reveals how these approaches may lead, in practice, to unjust attribution of responsibility, dilution of accountability, or barriers to effective patient redress. Building on this analysis, the article argues for a context-sensitive, factor-oriented normative framework for allocating responsibility in AI-assisted diagnosis. Rather than presuming responsibility to rest with any single actor, responsibility attribution should take into account key variables such as the degree of algorithmic explainability, the extent of AI’s practical influence within clinical workflows, the system’s validation and regulatory status, the maturity of professional guidelines, the nature of the error involved, and whether patients were adequately informed of AI involvement.
Future issues will be listed here as they are published.