Peripheral exudative hemorrhagic chorioretinopathy (PEHCR) is a peripheral retinal disease characterized by subretinal hemorrhage and/or subretinal pigment epithelial hemorrhage or exudation. It is often misdiagnosed as age-related macular degeneration, polypoidal chorioretinopathy or choroidal melanoma. With the development of multimodal imaging, PEHCR has different features under different examinations, such as B-scan ultrasound, fluorescein fundus angiography, optical coherence tomography and so on, which contributes to differention from other diseases. Clinical treatments for the disease include intravitreal injection of retinal photocoagulation therapy, anti-vascular endothelial growth factor, pars plana vitrectomyand so on, but there is still no universal consensus. In order to gain a deeper understanding of the clinical features, treatment options and prognosis of PEHCR, minimize missed diagnoses and misdiagnoses, and improve treatment efficiency, further research is required.
Central serous chorioretinitis (CSC) is a kind of choroidal retinopathy characterized by choroidal vasodilatation and hyperpermeability, retinal pigment epithelial cell lesions and serous retinal detachment. Various imaging examinations and imaging techniques have been used to describe the characteristics of the retina and choroid. Fundus manifestations of different types of CSC has both generality, and have their respective characteristic. The classification of CSC and its differentiation from other diseases including the choroidal neovascularization and pachychoroidopathy spectrum depending on varieties of fundus imaging techniques. The current study aims to review the various performance characteristics of CSC especially for chronic CSC with multimodal imaging and the current research progress, so as to provide reference for ophthalmologists to more comprehensively and intuitively understand the clinical characteristics and potential pathogenesis of CSC, and also to provide basis for multimodal imaging assisted diagnosis and treatment.
Cardiovascular diseases is the leading cause of threat to human life and health worldwide. Early risk assessment, timely diagnosis, and prognosis evaluation are critical to the treatment of cardiovascular diseases. Currently, the evaluation of diagnosis and prognosis of cardiovascular diseases mainly relies on imaging examinations such as coronary CT and coronary angiography, which are expensive, time-consuming, partly invasive, and require high professional competence of the operator, making it difficult to promote in the community or in areas where medical resources are scarce. The fundus microcirculation is a part of the human microcirculation and has similar embryological origins and physiopathological features to cardiovascular circulation. Several studies have revealed fundus imaging biomarkers associated with cardiovascular diseases, and developed and validated intelligent diagnosis and treatment models for cardiovascular diseases based on fundus imaging data. Fundus imaging is expected to be an important adjunct to cardiovascular disease diagnosis and treatment given its noninvasive and convenient nature. The purpose of this review is to summarize the current research status, challenges, and future prospects of the application of artificial intelligence based on multimodal fundus imaging data in cardiovascular disease diagnosis and treatment.
Currently, the development of deep learning-based multimodal learning is advancing rapidly, and is widely used in the field of artificial intelligence-generated content, such as image-text conversion and image-text generation. Electronic health records are digital information such as numbers, charts, and texts generated by medical staff using information systems in the process of medical activities. The multimodal fusion method of electronic health records based on deep learning can assist medical staff in the medical field to comprehensively analyze a large number of medical multimodal data generated in the process of diagnosis and treatment, thereby achieving accurate diagnosis and timely intervention for patients. In this article, we firstly introduce the methods and development trends of deep learning-based multimodal data fusion. Secondly, we summarize and compare the fusion of structured electronic medical records with other medical data such as images and texts, focusing on the clinical application types, sample sizes, and the fusion methods involved in the research. Through the analysis and summary of the literature, the deep learning methods for fusion of different medical modal data are as follows: first, selecting the appropriate pre-trained model according to the data modality for feature representation and post-fusion, and secondly, fusing based on the attention mechanism. Lastly, the difficulties encountered in multimodal medical data fusion and its developmental directions, including modeling methods, evaluation and application of models, are discussed. Through this review article, we expect to provide reference information for the establishment of models that can comprehensively utilize various modal medical data.
Objective To improve hand hygiene executive ability of healthcare workers in medical institutions in Anhui Province by multi-modal interventions with the administrative intervention as the guide. Methods The PDCA management mode was adopted in a step-by-step implementation of plan, implementation, inspection, improvement, and effectiveness evaluation in Anhui Province from April 2014 to December 2016. The management indicators of hand hygiene before and after the intervention in 1 353 hospitals were investigated and evaluated. Results The overall evaluation of the hand hygiene at the end of the implemention showed that 85.29% (58/68) of the tertiary hospitals, 84.07% (227/270) of the second-class hospitals and 66.63% (595/893) of the primary-level hospitals had well-equipped hand hygiene facilities. About 92.65% (63/68) of the tertiary hospitals, 100.00% (270/270) of the second-class hospitals and 50.06% (447/893) of the primary-level hospitals had staff training of hand hygiene knowledge. The compliance of hand hygiene before and after intervention increased from 36.68% to 61.93%, the correct rate of hand washing increased from 37.60% to 89.28%, the awareness rate of related knowledge increased from 41.20% to 86.07%, and the dosage of hand disinfectant increased from 2.59 mL to 7.10 mL. Conclusion To take multi-model interventions with the administrative intervention as the guide, can effectively improve the quality of hand hygiene management and the executive force.
Electrocardiogram (ECG) signal is an important basis for the diagnosis of arrhythmia and myocardial infarction. In order to further improve the classification effect of arrhythmia and myocardial infarction, an ECG classification algorithm based on Convolutional vision Transformer (CvT) and multimodal image fusion was proposed. Through Gramian summation angular field (GASF), Gramian difference angular field (GADF) and recurrence plot (RP), the one-dimensional ECG signal was converted into three different modes of two-dimensional images, and fused into a multimodal fusion image containing more features. The CvT-13 model could take into account local and global information when processing the fused image, thus effectively improving the classification performance. On the MIT-BIH arrhythmia dataset and the PTB myocardial infarction dataset, the algorithm achieved a combined accuracy of 99.9% for the classification of five arrhythmias and 99.8% for the classification of myocardial infarction. The experiments show that the high-precision computer-assisted intelligent classification method is superior and can effectively improve the diagnostic efficiency of arrhythmia as well as myocardial infarction and other cardiac diseases.
The risk prediction of paroxysmal atrial fibrillation (PAF) is a challenge in the field of biomedical engineering. This study integrated the advantages of machine learning feature engineering and end-to-end modeling of deep learning to propose a PAF risk prediction method based on multimodal feature fusion. Additionally, the study utilized four different feature selection methods and Pearson correlation analysis to determine the optimal multimodal feature set, and employed random forest for PAF risk assessment. The proposed method achieved accuracy of (92.3 ± 2.1)% and F1 score of (91.6 ± 2.9)% in a public dataset. In a clinical dataset, it achieved accuracy of (91.4 ± 2.0)% and F1 score of (90.8 ± 2.4)%. The method demonstrates generalization across multi-center datasets and holds promising clinical application prospects.
Objective To evaluate the clinical features of acute macular neuroretinopathy (AMN) associated with COVID-19. MethodsA retrospective case series studies. A total of 12 eyes of 8 patients diagnosed of AMN associated with COVID-19 at Peking University People’s Hospital from December 5, 2022 to January 5, 2023 were included. Of the 8 patients, 2 were male (4 eyes) and 6 were female (8 eyes), with an average age of (29.38±8.60) years. All patients underwent best-corrected visual acuity (BCVA), spectral-domain optical coherence tomography (OCT), and infra-red fundus photography (IR). After definite diagnosis, the patients were given symptomatic treatment such as local vasodilation, anti-inflammatory and systemic circulation improvement and nutritional nerve. Follow-up time was 21-30 days weeks. Clinical manifestations, OCT and IR image characteristics, and treatment outcomes were retrospectively analyzed. ResultsThe time from diagnosis of COVID-19 to the onset of ocular symptoms was (3.00±0.93) days. Among 12 eyes, 6 had complaints of paracentral scotoma, with 2 of them accompanied by visual acuity loss; and 6 had complaints of dark shadows in the vision, with 2 of them accompanied by visual acuity loss. At the initial examination, 2 eyes had a BCVA of less than 0.05, 2 eyes had a BCVA between 0.4 and 0.6, and 8 eyes had a BCVA between 0.8 and 1.0. At the last follow-up, visual symptoms improved in 7 eyes and remained unchanged in 5 eyes. Fundus color photography showed reddish-brown lesions in the macular area. Spectral-domain OCT revealed localized thickening and strong reflection of the outer plexiform layer (OPL) in the macular area, patchy strong reflections in the outer nuclear layer (ONL), and varying degrees of local discontinuity in the adjacent external limiting membrane, ellipsoid zone/interdigitation zone (EZ/IZ), with reduced local reflection in the adjacent retinal pigment epithelium layer in 2 eyes. The strong reflection area of the ONL on corresponding structural OCT was observed more clearly as a lesion range with strong reflection on en-face OCT. The incomplete structure of the EZ/IZ band was observed more clearly as a lesion range with weak reflection on en-face OCT. IR showed several clear-bordered and weakly reflecting lesions at the center of the macula, with the tip pointing to the fovea. ConclusionsAMN associated with COVID-19 tends to occur in young females. The OCT findings of AMN are characterized by strong reflections in the OPL and ONL, and lesion ranges can be observed more clearly at different levels using en-face OCT. The lesions on IR appear as weak reflections.
ObjectiveTo observe the clinical and imaging characteristics of acute idiopathic macular degeneration (AIM).MethodsA retrospective clinical study. From March 2016 to January 2018, 5 eyes (5 AIM patients) in The Second People's Hospital of Yunnan Province were included in the study. Among them, there were 4 males (4 eyes) and 1 female (1 eye); all patients were monocular with the average age of 34.2 years. The course of illness from onset of symptoms to treatment was 4-22 days. All affected eyes were examined by BCVA, fundus color photography, OCT, FAF, and FFA. Among 5 eyes, 1 eye with optic disc vasculitis was given oral glucocorticoid treatment; 4 eyes were not interfered after the diagnosis. ResultsThe follow-up time was 6 months. During follow-up, BCVA, fundus color photography, and OCT examination were performed. The results were all a sudden decrease in monocular vision, accompanied by visual distortion or central dark spots. At the first visit, the BCVA was 0.1, 0.2, 0.2, 0.05, and 0.5; at the last follow-up, the BCVA of the affected eye was 0.8, 0.6, 0.5, 0.5, and 1.0, respectively. Fundus color photography showed that at the first diagnosis, all the affected eyes showed irregular round yellow-white lesions in the macular area, including 1 eye with small patches of hemorrhage and 1 eye with pseudopyous changes in the macular area. Two to three weeks after the initial diagnosis, the yellowish-white lesions and bleeding in the macular area were basically absorbed. The center of the lesion showed weak pseudopod-like fluorescence, and the surrounding area was surrounded by strong fluorescence in FAF examination. The irregular and strong fluorescence in the early macular area and accumulation of late fluorescein in FFA examination. One eye was receivied glucocorticoid therapy. The upper layer of the retinal nerve in the macular area was detached, and the inferior space showed focal strong reflective material in 3 eyes in OCT examination. At the first diagnosis, the retinal neuroepithelial layer was detached, the top of the RPE layer was irregular with strong reflective material, and the structure of the ellipsoid zone and the chimera zone was unclear; as the course of the disease prolonged, the outer retinal structure recovered.ConclusionsAIM is characterized by inflammatory exudative changes in the outer layer of the retina in the macular area; FFA is characterized by strong subretinal disc-like fluorescence or multifocal weak fluorescence in the macular area; OCT mainly manifests as neuroepithelial detachment and changes in the outer retina and RPE, The structure can be restored by itself.
Objective To explore the research progress of the multimodal clinical support system (CSS). Methods With recognized development and operation of the multi-model CSS, and compared to the traditional CSS, to explore the research progress of the multimodal CSS. Results Based on the realization of the concept, purpose and characteristics of the multimodal CSS, it has been known that the international research progress of the multimodal CSS. Conclusion The developing and evolving of the CSS model have offered a new assist to the multi-disciplinary treatment model, and have enhanced the improving system associated with the practice of evidence-based medicine. However, the application of clinical support system program (CSSP) in our country still needs more research.