Artificial intelligent (AI) systems for clinical-decision support are an important tool in clinical routine. It has become a crucial diagnostic tool with adequate reliability and interpretability in disease diagnosis and monitoring. Undoubtedly, these models are faced with insufficient data challenges for training, which often directly determines the model’s performance. In order word, insufficient data for model training leads to inefficiency in the model built. To overcome this problem, we propose an AI-driven model by transfer learning in accurate diagnosis for medical decision support. Our approach leverages the shortage of data with a pretrained model by training the neural network with a fraction of the new dataset. For this purpose, we utilized the VGG19 network as the backbone network to support our model in integrating known features with the newly learned features for accurate diagnosis and decision making. Integrating this trained model speeds up the training phase and improve the performance of the proposed model. Experimental results show that the proposed model is effective and efficient in diagnosing different medical diseases. As such, we anticipated that this diagnosis tool will ultimately aid in facilitating early treatment of these treatable diseases, which will improve clinical out-comes.
23Mutations in the retinitis pigmentosa GTPase regulator (RPGR) gene, are the major cause of 24 X-linked retinitis pigmentosa (RP). Herein we used whole-exome sequencing to screen 25 possible novel RPGR mutations in RP patients, and identified a novel missense mutation 26 E585K in a patient with early onset but slow disease progression, and a frameshift deletion 27 E998Gfs*78 in a patient with RP sine pigmento and high myopia. Intriguingly, bioinformatic 28 analysis indicated that E585K probably affected RPGR RNA splicing instead of the protein 29 sequence directly. Mini-gene assays in 293T cells revealed that splicing events of the E585K 30 mutant were found to be also exist in wildtype, but with a shifted pattern. In the E585K mini-31 gene usage of an upstream alternative 5′ splice site (5′ ss) of exon 14 was enhanced, and other 32 splicing events were suppressed, including the canonical 5′ ss of exon 14, skipping of exon 33 14/15 and retention of intron 14. As a result, RPGR splicing products of the E585K mini-gene 34 were predominated by transcripts containing a 4-bp deletion, with a small fraction of in-frame 35 transcripts containing a retended intron 14, which might explain the slow disease progression 36 in the patient carrying the mutation. RNA-Seq analysis further confirmed existence of these 37 splicing events in endogenous RPGR RNA in human retina, pointing to compromised splicing 38 diversity in the E585K mutant. Our findings thus added to the understanding of genotype-39 phenotype correlation in RP, and suggested that compromised RPGR splicing diversity might 40 play a role in molecular mechanism of the disease. 41
Purpose. To report the ratio of repeat-to-initial keratoplasty among patients who had underwent therapeutic keratoplasty for microbial keratitis in Southern China and to investigate the characteristics and risk factors of repeat keratoplasty. Methods. A retrospective and inclusive review of the clinical records of patients who had received therapeutic keratoplasty for microbial keratitis, at Zhongshan Ophthalmic Center during December 2012 to January 2018, was performed. Patients who suffered coexistent endophthalmitis or underwent keratoplasty combined with other surgeries were excluded. Data on clinical characteristics of all eligible patients were collected. Results. A total of 447 patients were identified. Their mean age was 48.7 ± 15.5 years, and 290 (64.9%) were male. Out of the 447 patients, 18 (4.0%) received repeat keratoplasty. Their mean age was 45.9 ± 11.3 years, and 14 (77.8%) were male. The most common indication of repeat keratoplasty (12/18) was refractory infectious keratitis. Most of the patients (15/18, 83.3%) received the second keratoplasty within 12 months after the initial keratoplasty. Factors, including age, gender, initial causative organism, presence of initial corneal perforation, ocular comorbidities, and surgical procedures were not found statistically significantly different between patients who received and not received repeat keratoplasty. Conclusion. The ratio of repeat-to-initial keratoplasty for therapeutic keratoplasty is low, compared to a failure rate of the initial grafts of over 50% reported in previous studies. The low ratio and the most common indication of repeat keratoplasty, refractory infectious keratitis, reflect caution for performing regrafts in such patients.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.