Practical teaching is an important link in enhancing the professional abilities and professional qualities of medical physics talents. This paper explores a new practical teaching model for cultivating medical physics talents based on an X-ray diagnosis and treatment scientific research and innovation platform. By analyzing the advantages of the X-ray diagnosis and treatment platform, combined with student research projects and ability development, the training experience is summarized. The study shows that students have developed programming design and analysis, experimental design and hands-on skills, scientific research thinking and innovation, and social collaboration and coordination abilities through direct participation in the development and application of the X-ray diagnosis and treatment platform. The proposed practical teaching model can mutually reinforce the teaching and scientific research, providing new ideas for cultivating high-quality medical physics students’ innovative practical abilities.

X-ray luminescence computed tomography (XLCT) is an advanced imaging technology that combines X-ray excitation and optical detection, with wide applications in medical imaging, materials science, biomedical engineering, and other fields. With the continuous development of this technology, its research achievements have not only promoted the progress of scientific research but also provided new resources and methods for higher education, especially for the teaching of medical physics, medical imaging, and related disciplines. This article explores how XLCT research can feed back into medical physics teaching, including promoting interdisciplinary integration, optimizing the experimental teaching system, promoting problem-based learning (PBL) mode, and improving the assessment and evaluation system. The aim is to provide theoretical support and practical solutions for teaching reform in universities. The analysis of practical effects shows that XLCT research can effectively feed back into university teaching, enhance students' innovation ability and practical level, and provide strong support for cultivating high-quality composite talents.

Colorectal cancer, a malignant lesion of the intestines, significantly affects human health and life, emphasizing the necessity of early detection and treatment. Accurate segmentation of colorectal cancer regions directly impacts subsequent staging, treatment methods, and prognostic outcomes. While colonoscopy is an effective method for detecting colorectal cancer, its data collection approach can cause patient discomfort. To address this, current research utilizes Computed Tomography (CT) imaging; however, conventional CT images only capture transient states, lacking sufficient representational capability to precisely locate colorectal cancer. This study utilizes enhanced CT images, constructing a deep feature network from the arterial, portal venous, and delay phases to simulate the physician’s diagnostic process and achieve accurate cancer segmentation. The innovations include: 1) Utilizing portal venous phase CT images to introduce a context-aware multi-scale aggregation module for preliminary shape extraction of colorectal cancer. 2) Building an image sequence based on arterial and delay phases, transforming the cancer segmentation issue into an anomaly detection problem, establishing a pixel-pairing strategy, and proposing a colorectal cancer segmentation algorithm using a Siamese network. Experiments with 84 clinical cases of colorectal cancer enhanced CT data demonstrated an Area Overlap Measure of 0.90, significantly better than Fully Convolutional Networks (FCNs) at 0.20. Future research will explore the relationship between conventional and enhanced CT to further reduce segmentation time and improve accuracy.