ObjectiveTo conduct a meta-analysis comparing the accuracy of artificial intelligence (AI)-assisted diagnostic systems based on 18F-fluorodeoxyglucose PET/CT (18F-FDG PET/CT) and structural MRI (sMRI) in the diagnosis of Alzheimer's disease (AD). MethodsOriginal studies dedicated to the development or validation of AI-assisted diagnostic systems based on 18F-FDG PET/CT or sMRI for AD diagnosis were retrieved from the Web of Science, PubMed, and Embase databases. Studies meeting the inclusion criteria were collected, and the risk of bias and clinical applicability of the included studies were assessed using the PROBAST checklist. The pooled sensitivity, specificity, and area under the summary receiver operating characteristic (SROC) curve (AUC) were calculated using a bivariate random-effects model. ResultsTwenty-six studies met the inclusion criteria, yielding a total of 38 2×2 contingency tables related to diagnostic performance. Specifically, 24 contingency tables were based on 18F-FDG PET/CT to distinguish AD patients from normal cognitive (NC) controls, and 14 contingency tables were based on sMRI for the same purpose. The meta-analysis results showed that for 18F-FDG PET/CT, the AI-assisted diagnostic systems had a pooled sensitivity, specificity, and SROC-AUC of 89% (95%CI 88% to 91%), 93% (95%CI 91% to 94%), and 0.96 (95%CI 0.93 to 0.97), respectively. For sMRI, the AI-assisted diagnostic systems had a pooled sensitivity, specificity, and SROC-AUC of 88% (95%CI 85% to 90%), 90% (95%CI 87% to 92%), and 0.94 (95%CI 0.92 to 0.96), respectively. ConclusionAI-assisted diagnostic systems based on either 18F-FDG PET/CT or sMRI demonstrated similar performance in the diagnosis of AD, with both showing high accuracy.
China is facing the peak of an ageing population, and there is an increase in demand for intelligent healthcare services for the elderly. The metaverse, as a new internet social communication space, has shown infinite potential for application. This paper focuses on the application of the metaverse in medicine in the intervention of cognitive decline in the elderly population. The problems in assessment and intervention of cognitive decline in the elderly group were analyzed. The basic data required to construct the metaverse in medicine was introduced. Moreover, it is demonstrated that the elderly users can conduct self-monitoring, experience immersive self-healing and health-care through the metaverse in medicine technology. Furthermore, we proposed that it is feasible that the metaverse in medicine has obvious advantages in prediction and diagnosis, prevention and rehabilitation, as well as assisting patients with cognitive decline. Risks for its application were pointed out as well. The metaverse in medicine technology solves the problem of non-face-to-face social communication for elderly users, which may help to reconstruct the social medical system and service mode for the elderly population.
With the development of society and the progress of technology, artificial intelligence (AI) and big data technology have penetrated into all walks of life in social production and promoted social production and lifestyle greatly. In the medical field, the applications of AI, such as AI-assisted diagnosis and treatment, robots, medical imaging and so on, have greatly promoted the development and transformation of the entire medical industry. At present, with the support of national policy, market, and technology, we should seize the opportunity of AI development, so as to build the first-mover advantage of AI development. Of course, the development and challenges are coexisted. In the future development process, we should objectively analyze the gap between our country and developed countries, think about the unfavorable factors such as AI chips and data problems, and extend the application and service of AI and big data to all links of medical industry, integrate with clinic fully, so as to better promote the further development of AI medicine treatment in China.
ObjectiveTo apply the multi-modal deep learning model to automatically classify the ultra-widefield fluorescein angiography (UWFA) images of diabetic retinopathy (DR). MethodsA retrospective study. From 2015 to 2020, 798 images of 297 DR patients with 399 eyes who were admitted to Eye Center of Renmin Hospital of Wuhan University and were examined by UWFA were used as the training set and test set of the model. Among them, 119, 171, and 109 eyes had no retinopathy, non-proliferative DR (NPDR), and proliferative DR (PDR), respectively. Localization and assessment of fluorescein leakage and non-perfusion regions in early and late orthotopic images of UWFA in DR-affected eyes by jointly optimizing CycleGAN and a convolutional neural network (CNN) classifier, an image-level supervised deep learning model. The abnormal images with lesions were converted into normal images with lesions removed using the improved CycleGAN, and the difference images containing the lesion areas were obtained; the difference images were classified by the CNN classifier to obtain the prediction results. A five-fold cross-test was used to evaluate the classification accuracy of the model. Quantitative analysis of the marker area displayed by the differential images was performed to observe the correlation between the ischemia index and leakage index and the severity of DR. ResultsThe generated fake normal image basically removed all the lesion areas while retaining the normal vascular structure; the difference images intuitively revealed the distribution of biomarkers; the heat icon showed the leakage area, and the location was basically the same as the lesion area in the original image. The results of the five-fold cross-check showed that the average classification accuracy of the model was 0.983. Further quantitative analysis of the marker area showed that the ischemia index and leakage index were significantly positively correlated with the severity of DR (β=6.088, 10.850; P＜0.001). ConclusionThe constructed multimodal joint optimization model can accurately classify NPDR and PDR and precisely locate potential biomarkers.
The development of the fifth generation mobile networks (5G) technology has brought great breakthroughs and challenges to clinical medicine and medical education. In the context of “5G + medicine”, the development of telemedicine, emergency rescue, intelligent analysis and diagnosis has opened up new horizons for clinical medicine. Facing the constant impact of high technology, the focus of medical education should be on the cultivation of students’ integrated medical view, critical thinking, communication abilities and skills, and creativity. The “5G + education” model will be presented by means of virtual reality, artificial intelligence, cloud computing and other technologies, providing a new direction for the development of medical education. This article summarizes the key points and prospects of medical education under 5G technology in order to provide a reference for the field of medical education to adapt to the changes in the 5G era.
For the past few years, artificial intelligence (AI) technology has developed rapidly and has become frontier and hot topics in medical research. While the deep learning algorithm based on artificial neural networks is one of the most representative tool in this field. The advancement of ophthalmology is inseparable from a variety of imaging methods, and the pronounced convenience and high efficiency endow AI technology with promising applications in screening, diagnosis and follow-up of ophthalmic diseases. At present, related research on ophthalmologic AI technology has been carried out in terms of multiple diseases and multimodality. Many valuable results have been reported aiming at several common diseases of ophthalmology. It should be emphasized that ophthalmic AI products are still faced with some problems towards practical application. The regulatory mechanism and evaluation criteria have not yet integrated as a standardized system. There are still a number of aspects to be optimized before large-scale distribution in clinical utility. Briefly, the innovation of ophthalmologic AI technology is attributed to multidisciplinary cooperation, which is of great significance to China's public health undertakings, and will be bound to benefit patients in future clinical practice.
High myopia has become a global public health issue, posing a significant threat to visual health. There are still some problems in the process of diagnosis and treatment, including the definition of high myopia and pathological myopia, opportunities and challenges of artificial intelligence in the diagnosis and treatment system, domestic and international collaboration in the field of high myopia, the application of genetic screening in children with myopia and high myopia patients, and the exploration of new treatment methods for high myopia. Nowadays, myopia and high myopia show the characteristics of early onset age and sharp rise in prevalence, and gradually become the main cause of low vision and irreversible blindness in young and middle-aged people. Therefore, it is of great significance to accurately define high myopia and pathological myopia, combine artificial intelligence and other methods for screening and prevention, promote cooperation in different fields, strengthen gene screening for early-onset myopia and adopt new and effective ways to treat it.
ObjectiveTo compare the clinical application of empirical thoracoscopic segmentectomy and precise segmentectomy planned by artificial intelligence software, and to provide some reference for clinical segmentectomy. MethodsA retrospective analysis was performed on the patients who underwent thoracoscopic segmentectomy in our department from 2019 to 2022. The patients receiving empirical thoracoscopic segmentectomy from January 2019 to September 2021 were selected as a group A, and the patients receiving precise segmentectomy from October 2021 to December 2022 were selected as a group B. The number of preoperative Hookwire positioning needle, proportion of patients meeting oncology criteria, surgical time, intraoperative blood loss, postoperative chest drainage time, postoperative hospital stay, and number of patients converted to thoracotomy between the two groups were compared. Results A total of 322 patients were collected. There were 158 patients in the group A, including 56 males and 102 females with a mean age of 56.86±8.82 years, and 164 patients in the group B, including 55 males and 109 females with a mean age of 56.69±9.05 years. All patients successfully underwent thoracoscopic segmentectomy, and patients whose resection margin did not meet the oncology criteria were further treated with extended resection or even lobectomy. There was no perioperative death. The number of positioning needles used for segmentectomy in the group A was more than that in the group B [47 (29.7%) vs. 9 (5.5%), P<0.001]. There was no statistical difference in the number of positioning needles used for wedge resection between the two groups during the same period (P=0.572). In the group A, the nodule could not be found in the resection target segment in 3 patients, and the resection margin was insufficient in 10 patients. While in the group B, the nodule could not be found in 1 patient, and the resection margin was insufficient in 3 patients. There was a statistical difference between the two groups [13 (8.2%) vs. 4 (2.4%), P=0.020]. There was no statistical difference between the two groups in terms of surgical time, intraoperative blood loss, duration of postoperative thoracic drainage, postoperative hospital stay, or conversion to open chest surgery (P＞0.05). Conclusion Preoperative surgical planning performed with the help of artificial intelligence software can effectively guide the completion of thoracoscopic anatomical segmentectomy. It can effectively ensure the resection margin of pulmonary nodules meeting the oncological requirements and significantly reduce the number of positioning needles of pulmonary nodules.
Objective To investigating the safety and accuracy of artificial intelligence (AI) assisted automatic planning of pedicle screws parallel to sagittal plane for C1. Methods The subjects who completed cervical CT scan in Zigong Fourth People’s Hospital btween January 2020 and December 2023 were selected. The subjects who completed cervical CT scan were randomly divided into two groups using a random number table method. Among them, 80% were used as the training model (training group), and 20% were used as the validation model (validation group). The original cervical CT data of the training group were imported into ITK-SNAP software to mark the feature points. Four feature points were selected. In order to obtain the weighted function model of the four feature points, training group were trained with the spatial key point location algorithm. pedicle trajectory based on the four key points obtained. Finally, the algorithm was compiled to form a visual interface, and imported into the verification group of annular vertebral CT data to calculate the pedicle screw trajectory. Results A total of 500 patients were included. Among them, there were 400 cases in the training group and 100 cases in the validation group. The average positioning error of spatial key points is (0.47±0.16) mm. The average distance between the planned pedicle screw center line and the internal edge of the pedicle was (2.86±0.12) mm. Pedicle screw placement parallel to the sagittal plane and 3D display can be safely performed for the C1 pedicle that is large enough to accommodate a 3.5 mm diameter screw without cortical breakthrough. Conclusions For pedicle screw planning parallel to the sagittal plane in C1, training based on the spatial positioning algorithm of anterior and posterior tubercles and bilateral tangential points can obtain a safe and accurate pedicle screw trajectory. It provides theoretical basis for orthopedic robot automatic screw placement. For vertebral bodies with narrow or deformed pedicles, further expansion of the training data is needed to expand the adaptive range and improve the accuracy of the algorithm.
Ophthalmic imaging examination is the main basis for early screening, evaluation and diagnosis of eye diseases. In recent years, with the improvement of computer data analysis ability, the deepening of new algorithm research and the popularization of big data platform, artificial intelligence (AI) technology has developed rapidly and become a hot topic in the field of medical assistant diagnosis. The advantage of AI is accurate and efficient, which has great application value in processing image-related data. The application of AI not only helps to promote the development of AI research in ophthalmology, but also helps to establish a new medical service model for ophthalmic diagnosis and promote the process of prevention and treatment of blindness. Future research of ophthalmic AI should use multi-modal imaging data comprehensively to diagnose complex eye diseases, integrate standardized and high-quality data resources, and improve the performance of algorithms.