ObjectiveTo observe the multimodal imaging characteristics of tamoxifen retinopathy. MethodsA retrospective case study. From January 2019 to December 2021, 4 patients (8 eyes) with tamoxifen retinopathy diagnosed in Tangshan Eye Hospital were included in the study. All patients were female, with sick binoculus. The age was 59.5±4.6 years. After breast cancer resection, tamoxifen 20 mg/d was taken orally consecutively, including 1, 1, and 2 cases who took tamoxifen orally for 5, 7, and ≥10 years. All eyes were examined by fundus color photography, optical coherence tomography (OCT), OCT angiography (OCTA), fundus fluorescein angiography (FFA), and fundus autofluorescence (AF). The multi-mode image features of the fundus of the affected eyes were observed. ResultsThe yellow white dot crystal like material deposition in the macular area was observed in all eyes. In fundus AF examination, macular area showed patchy strong AF. FFA examination showed telangiectasia and fluorescein leakage in macular area at late stage. OCT showed that punctate strong reflexes could be seen between the neuroepithelial layers in the macular region with the formation of a space between the neuroepithelial layers, the interruption of the elliptical zone (EZ), and the formation of a hole in the outer lamella including 4, 5 and 3 eyes; The thickness of ganglion cells in macular region decreased in 7 eyes. OCTA showed that the blood flow density of the superficial retinal capillary plexus around the arch ring was decreased, and the retinal venules were dilated in 2 eyes; Deep capillary plexus (DCP) showed telangiectasia. ConclusionDeposition of yellowish white dot like crystals can be seen in the macular region of tamoxifen retinopathy; dotted strong reflex between neuroepithelial layers, cavity formation, thinning of ganglion cell layer, EZ middle fissure and outer lamellar fissure; DCP capillaries and venules around the arch were dilated; telangiectasia in macular region; flaky strong AF in macular region.
Objective To observe the multimodal imaging characteristics of the eyes in patients with presumed tuberculous retinal vasculitis. Methods A retrospective case series study. A total of 15 patients (22 eyes) diagnosed with presumed tuberculous retinal vasculitis and receiving anti-tuberculosis treatment (ATT) effectively in Department of Ophthalmology, Subei People's Hospital Affiliated to Yangzhou University from January 2018 to April 2021 were included. Among them, there were 5 males and 10 females. Seven had bilateral involvement and 8 had unilateral involvement. The age was 49.3±11.1 years old. The best corrected visual acuity (BCVA), fundus colour photography, wide-angle fundus fluorescein angiography (FFA), and optical coherence tomography (OCT) were performed in all patients. Indocyanine green angiography (ICGA) was performed in 7 eyes. The BCVA examination was performed with the international standard visual acuity chart, which was converted into the logarithm of minimal angel resolution vision (logMAR). Systemic tuberculosis-related examinations included chest CT, serum T-spot, purified protein derivative and other tuberculosis-related tests. All patients were treated with systemic anti-tuberculosis therapy. The follow-up time was ＞12 months. The multimodal imaging characteristics for affected eyes. Nonparametric test was used to compare BCVA before and after treatment. ResultsThe retinal vessels of all the affected eyes were tortuously dilated, including 3 eyes with vascular white scabbard, 5 eyes with scattered bleeding point at the retina inculding 3 eyes walking along the vessels. The lesions were mainly distributed in the middle and periphery of the retina, and some of them involved the posterior pole; 12 eyes (54.5%, 12/22) with simple retinal vasculitis and 10 eyes (45.5%, 10/22) with retinal vasculitis complicated with choroiditis. Tuberculous retinal vasculitis showed different degrees of retinal vascular leakage on FFA, mainly retinal vein and capillary leakage, not involving arteries; 16 eyes (72.7%, 16/22) of retinal vasculitis showed peripheral occlusive retinal vasculitis and 4 eyes (18.2%, 4/22) were associated with retinal neovascularization. In 10 eyes with choroiditis, there were multiple focal choroiditis lesions of different sizes under the retina. Of the 7 eyes examined by ICGA, the choroidal inflammatory lesions showed hypofluorescent dark dots (HDD) in 5 eyes (71.4%,5/7), showing HDDs of different sizes, most of which were distributed in the posterior pole and middle periphery. In 10 eyes with retinal vasculitis complicated with choroiditis after ATT, the accumulation of hyper-reflective substances above and below the retinal pigment epithelium layer of the retina was gradually absorbed, but not completely disappeared, and most of the disorders of retinal structure could not be recovered. The average logMAR visual acuity was 0.61±0.57 before treatment and 0.36±0.55 after treatment. The BCVA after treatment was significantly higher than that before treatment (Z=－3.102, P＜0.01). ConclusionsPeripheral occlusive retinal vasculitis is the most common manifestation of tuberculous retinal vasculitis in FFA, which may be accompanied by focal choroidal inflammatory lesions. Wide-angle FFA and ICGA are more important in the diagnosis of tuberculous retinal vasculitis. OCT can be used for monitoring the changes of inflammation.
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.
In recent years, the incidence of thyroid diseases has increased significantly and ultrasound examination is the first choice for the diagnosis of thyroid diseases. At the same time, the level of medical image analysis based on deep learning has been rapidly improved. Ultrasonic image analysis has made a series of milestone breakthroughs, and deep learning algorithms have shown strong performance in the field of medical image segmentation and classification. This article first elaborates on the application of deep learning algorithms in thyroid ultrasound image segmentation, feature extraction, and classification differentiation. Secondly, it summarizes the algorithms for deep learning processing multimodal ultrasound images. Finally, it points out the problems in thyroid ultrasound image diagnosis at the current stage and looks forward to future development directions. This study can promote the application of deep learning in clinical ultrasound image diagnosis of thyroid, and provide reference for doctors to diagnose thyroid disease.