Intelligent diagnostic model for pterygium by combining attention mechanism and MobileNetV2.

AIM: To evaluate the application of an intelligent diagnostic model for pterygium. METHODS: For intelligent diagnosis of pterygium, the attention mechanisms-SENet, ECANet, CBAM, and Self-Attention-were fused with the lightweight MobileNetV2 model structure to construct a tri-classification model. The study used 1220 images of three types of anterior ocular segments of the pterygium provided by the Eye Hospital of Nanjing Medical University. Conventional classification models-VGG16, ResNet50, MobileNetV2, and EfficientNetB7-were trained on the same dataset for comparison. To evaluate model performance in terms of accuracy, Kappa value, test time, sensitivity, specificity, the area under curve (AUC), and visual heat map, 470 test images of the anterior segment of the pterygium were used. RESULTS: The accuracy of the MobileNetV2+Self-Attention model with 281 MB in model size was 92.77%, and the Kappa value of the model was 88.92%. The testing time using the model was 9ms/image in the server and 138ms/image in the local computer. The sensitivity, specificity, and AUC for the diagnosis of pterygium using normal anterior segment images were 99.47%, 100%, and 100%, respectively; using anterior segment images in the observation period were 88.30%, 95.32%, and 96.70%, respectively; and using the anterior segment images in the surgery period were 88.18%, 94.44%, and 97.30%, respectively. CONCLUSION: The developed model is lightweight and can be used not only for detection but also for assessing the severity of pterygium.

Single-center retrospective study of the diagnostic value of double-balloon enteroscopy in Meckel's diverticulum with bleeding.

BACKGROUND: The study aimed to analyze the characteristic clinical manifestations of patients with intestinal disease Meckel's diverticulum (MD) complicated by digestive tract hemorrhage. Moreover, we aimed to evaluate the value of double-balloon enteroscopy (DBE) in MD diagnosis and the prognosis after laparoscopic diverticula resection. AIM: To evaluate the value of DBE in the diagnosis and the prognosis after laparoscopic diverticula resection for MD with bleeding. METHODS: The study retrospectively analyzed relevant data from 84 MD patients treated between January 2015 and March 2022 and recorded their clinical manifestations, auxiliary examination, and follow-up after laparoscopic resection of diverticula. RESULTS: (1) Among 84 MD patients complicated with hemorrhage, 77 were male, and 7 were female with an average age of 31.31 ± 10.75 years. The incidence was higher in men than in women of different ages; (2) Among the 84 MD patients, 65 (78.40%) had defecated dark red stools, and 50 (58.80%) had no accompanying symptoms during bleeding, indicating that most MD bleeding appeared a dark red stool without accompanying symptoms; (3) The shock index of 71 patients (85.20%) was < 1, suggesting that the blood loss of most MD patients was less than 20%-30%, and only a few patients had a blood loss of > 30%; (4) The DBE-positive rate was 100% (54/54), 99mTc-pertechnetate-positive scanning rate was 78% (35/45) compared with capsule endoscopy (36%) and small intestine computed tomography (19%). These results suggest that DBE and 99mTc-pertechnetate scans had significant advantages in diagnosing MD and bleeding, especially DBE was a highly precise examination method in MD diagnosis; (5) A total of 54 MD patients with hemorrhage underwent DBE examination before surgery. DBE endoscopy revealed many mucosal manifestations including normal appearance, inflammatory changes, ulcerative changes, diverticulum inversion, and nodular hyperplasia, with ulcerative changes being the most common (53.70%). This suggests that diverticular mucosal ulcer was the main cause of MD and bleeding; and (6) Laparoscopic dissection of diverticulae was performed in 76 patients, The patients who underwent postoperative follow-up did not experience any further bleeding. Additionally, follow-up examination of the 8 cases who had declined surgery revealed that 3 of them experienced a recurrence of digestive tract bleeding. These findings indicate that laparoscopic diverticula resection in MD patients complicated by bleeding had a favorable prognosis. CONCLUSION: Bleeding associated with MD was predominantly observed in male adolescents, particularly at a young age. DBE was a highly precise examination method in MD diagnosis. Laparoscopic diverticula resection effectively prevented MD bleeding and had a good prognosis.

Design, Synthesis, and Structure-Activity Relationship of Novel Pyridazinone-Based PARP7/HDACs Dual Inhibitors for Elucidating the Relationship between Antitumor Immunity and HDACs Inhibition.

Histone deacetylases (HDACs) inhibitors such as vorinostat (SAHA) has been used to treat hematologic malignancies (rather than solid tumors) and have been found to suppress the JAK/STAT, a critical signal pathway for antitumor immunity, while PARP7 inhibitor RBN-2397 could activate the type I interferons (IFN-I) pathway, facilitating downstream effects such as STAT1 phosphorylation and immune activation. To elucidate whether simultaneous inhibition of these two targets could interfere with these two signal pathways, a series of pyridazinone-based PARP7/HDACs dual inhibitors have been designed, synthesized, and evaluated in vitro and in vivo experiments. Compound 9l was identified as a potent and balanced dual inhibitor for the first time, exhibiting excellent antitumor capabilities both in vitro and in vivo. This suggests that 9l can be used as a valuable tool molecule for investigating the relationship between anticancer immunity and HDAC inhibition.

Design, synthesis and biological evaluation of novel 1,2,4a,5-tetrahydro-4H-benzo[b][1,4]oxazino[4,3-d][1,4]oxazine-based AAK1 inhibitors with anti-viral property against SARS-CoV-2.

Coronavirus entry into host cells hinges on the interaction between the spike glycoprotein of the virus and the cell-surface receptor angiotensin-converting enzyme 2 (ACE2), initiating the subsequent clathrin-mediated endocytosis (CME) pathway. AP-2-associated protein kinase 1 (AAK1) holds a pivotal role in this pathway, regulating CME by modulating the phosphorylation of the µ subunit of adaptor protein 2 (AP2M1). Herein, we report a series of novel AAK1 inhibitors based on previously reported 1,2,4a,5-tetrahydro-4H-benzo[b] [1,4]oxazino[4,3-d] [1,4]oxazine scaffold. Among 23 synthesized compounds, compound 12e is the most potent one with an IC50 value of 9.38 ± 0.34 nM against AAK1. The in vitro antiviral activity of 12e against SARS-CoV-2 was evaluated using a model involving SARS-CoV-2 pseudovirus infecting hACE2-HEK293 host cells. The results revealed that 12e was superior in vitro antiviral activity against SARS-CoV-2 entry into host cells when compared to SGC-AAK1-1 and LX9211, and its activity was comparable to that of a related and reference compound 8. Mechanistically, all AAK1 inhibitors attenuated AAK1-induced phosphorylation of AP2M1 threonine 156 and disrupted the direct interaction between AP2M1 and ACE2, ultimately inhibiting SARS-CoV-2 infection. Notably, compounds 8 and 12e exhibited a more potent effect in suppressing the phosphorylation of AP2M1 T156 and the interaction between AP2M1 and ACE2. In conclusion, novel AAK1 inhibitor 12e demonstrates significant efficacy in suppressing SARS-CoV-2 infection, and holds promise as a potential candidate for developing novel antiviral drugs against SARS-CoV-2 and other coronavirus infections.

Recent progress in discovery of novel AAK1 inhibitors: from pain therapy to potential anti-viral agents.

Adaptor associated kinase 1 (AAK1), a member of the Ark1/Prk1 family of Ser/Thr kinases, is a specific key kinase regulating Thr156 phosphorylation at the µ2 subunit of the adapter complex-2 (AP-2) protein. Due to their important biological functions, AAK1 systems have been validated in clinics for neuropathic pain therapy, and are being explored as potential therapeutic targets for diseases caused by various viruses such as Hepatitis C (HCV), Dengue, Ebola, and COVID-19 viruses and for amyotrophic lateral sclerosis (ALS). Centreing on the advances of drug discovery programs in this field up to 2023, AAK1 inhibitors are discussed from the aspects of the structure-based rational molecular design, pharmacology, toxicology and synthetic routes for the compounds of interest in this review. The aim is to provide the medicinal chemistry community with up-to-date information and to accelerate the drug discovery programs in the field of AAK1 small molecule inhibitors.

Artificial intelligence assisted pterygium diagnosis: current status and perspectives.

Pterygium is a prevalent ocular disease that can cause discomfort and vision impairment. Early and accurate diagnosis is essential for effective management. Recently, artificial intelligence (AI) has shown promising potential in assisting clinicians with pterygium diagnosis. This paper provides an overview of AI-assisted pterygium diagnosis, including the AI techniques used such as machine learning, deep learning, and computer vision. Furthermore, recent studies that have evaluated the diagnostic performance of AI-based systems for pterygium detection, classification and segmentation were summarized. The advantages and limitations of AI-assisted pterygium diagnosis and discuss potential future developments in this field were also analyzed. The review aims to provide insights into the current state-of-the-art of AI and its potential applications in pterygium diagnosis, which may facilitate the development of more efficient and accurate diagnostic tools for this common ocular disease.

Research on classification method of high myopic maculopathy based on retinal fundus images and optimized ALFA-Mix active learning algorithm.

AIM: To conduct a classification study of high myopic maculopathy (HMM) using limited datasets, including tessellated fundus, diffuse chorioretinal atrophy, patchy chorioretinal atrophy, and macular atrophy, and minimize annotation costs, and to optimize the ALFA-Mix active learning algorithm and apply it to HMM classification. METHODS: The optimized ALFA-Mix algorithm (ALFA-Mix+) was compared with five algorithms, including ALFA-Mix. Four models, including ResNet18, were established. Each algorithm was combined with four models for experiments on the HMM dataset. Each experiment consisted of 20 active learning rounds, with 100 images selected per round. The algorithm was evaluated by comparing the number of rounds in which ALFA-Mix+ outperformed other algorithms. Finally, this study employed six models, including EfficientFormer, to classify HMM. The best-performing model among these models was selected as the baseline model and combined with the ALFA-Mix+ algorithm to achieve satisfactory classification results with a small dataset. RESULTS: ALFA-Mix+ outperforms other algorithms with an average superiority of 16.6, 14.75, 16.8, and 16.7 rounds in terms of accuracy, sensitivity, specificity, and Kappa value, respectively. This study conducted experiments on classifying HMM using several advanced deep learning models with a complete training set of 4252 images. The EfficientFormer achieved the best results with an accuracy, sensitivity, specificity, and Kappa value of 0.8821, 0.8334, 0.9693, and 0.8339, respectively. Therefore, by combining ALFA-Mix+ with EfficientFormer, this study achieved results with an accuracy, sensitivity, specificity, and Kappa value of 0.8964, 0.8643, 0.9721, and 0.8537, respectively. CONCLUSION: The ALFA-Mix+ algorithm reduces the required samples without compromising accuracy. Compared to other algorithms, ALFA-Mix+ outperforms in more rounds of experiments. It effectively selects valuable samples compared to other algorithms. In HMM classification, combining ALFA-Mix+ with EfficientFormer enhances model performance, further demonstrating the effectiveness of ALFA-Mix+.

Design, Synthesis and Biological Evaluation of Histone Deacetylase Inhibitors Based on Pyrrolo[2,3-d]pyrimidine and Pyrrolo[2,3-b]pyridine Scaffolds.

Histone deacetylases (HDACs) are validated targets for the development of anticancer drugs in epigenetics. We have designed and synthesized a series of novel HDAC inhibitors based on pyrrolo[2,3-d]pyrimidine and pyrrolo[2,3-b]pyridine scaffolds. Compound B3 {(E)-3-(4-(((1-(7H-pyrrolo[2,3-d]pyrimidin-4-yl)piperidin-4-yl)amino)methyl)phenyl)-N-hydroxyacrylamide} exhibits potent inhibitory activity against HDACs 1, 2, 3, 6, and 8 with IC50 values of 5.2, 6.0, 8.8, 4.4, and 173.0 nM, respectively. It exhibited potent antiproliferative effects against three tumour cell lines (IC50 values of 0.13, 0.37, and 1.11 µM, against MV-4-11, K562, and WSU-DLCL-2 cells, respectively) with two- to sixfold improvement relative to suberoylanilide hydroxamic acid (SAHA). Mechanistic studies on WSU-DLCL-2 cells revealed that B3 exhibits anticancer effects through the induction of G0 /G1 -phase arrest and promotion of apoptosis. The results of this study warrant further investigation of this compound series for the treatment of hematological malignancy.

Explosive eye injuries: characteristics, traumatic mechanisms, and prognostic factors for poor visual outcomes.

BACKGROUND: Explosions can produce blast waves, high-speed medium, thermal radiation, and chemical spatter, leading to complex and compound eye injuries. However, few studies have comprehensively investigated the clinical features of different eye injury types or possible risk factors for poor prognosis. METHODS: We retrospectively reviewed all consecutive records of explosive eye injuries (1449 eyes in 1115 inpatients) in 14 tertiary referral hospitals in China over 12 years (between January 2008 and December 2019). Data on demographics, eye injury types, ocular findings, treatments, and factors affecting visual prognosis were extracted from a standardized database of eye injuries and statistically analyzed. RESULTS: Mechanical ocular trauma accounted for 94.00% of explosion-related eye injuries, among which intraocular foreign bodies (IOFBs) resulted in 55.17% of open globe injuries (OGIs) and contusion caused 60.22% of close globe injuries (CGIs). Proliferative vitreous retinopathy (PVR) was more common in perforating (47.06%) and IOFB (26.84%) than in penetrating (8.79%) injuries, and more common with laceration (24.25%) than rupture (9.22%, P < 0.01). However, no difference was observed between rupture and contusion. Ultimately, 9.59% of eyes were removed and the final vision was ≤ 4/200 in 45.82% of patients. Poor presenting vision [odds ratio (OR) = 5.789], full-thickness laceration of the eyeball ≥ 5 mm (OR = 3.665), vitreous hemorrhage (OR = 3.474), IOFB (OR = 3.510), non-mechanical eye injury (NMEI, OR = 2.622, P < 0.001), rupture (OR = 2.362), traumatic optic neuropathy (OR = 2.102), retinal detachment (RD, OR = 2.033), endophthalmitis (OR = 3.281, P < 0.01), contusion (OR = 1.679), ciliary body detachment (OR = 6.592), zone III OGI (OR = 1.940), and PVR (OR = 1.615, P < 0.05) were significant negative predictors for poor visual outcomes. CONCLUSIONS: Explosion ocular trauma has complex mechanisms, with multiple eyes involved and poor prognosis. In lethal level I explosion injuries, eyeball rupture is a serious condition, whereas contusion is more likely to improve. In level II injuries, IOFBs are more harmful than penetrating injuries, and level IV represents burn-related eye injuries. PVR is more associated with penetrating mechanisms than with OGI. Identifying the risk predictors for visual prognosis can guide clinicians in the evaluation and treatment of ocular blast injuries.

Recent progress in nickel-catalyzed carboboration of alkenes.

Alkenes represent one of the most useful building blocks for organic synthesis, owing to their abundance and versatile reactivity. Transition metal (Pd, Cu, Co, Ni, Fe, etc.) catalyzed difunctionalization of alkenes provides efficient access to substituted molecules from readily available alkenes by installing functional groups across their carbon-carbon double bonds. Particularly, Nickel-based catalytic complexes have attracted a great deal of attention. This is because they are prone to undergoing oxidative addition and slow ß-hydride elimination, and can access both two-electron and radical pathways. Numerous elegant Ni-catalyzed cross-coupling methods, e.g., (hetero)arylboration, alkenylboration, alkylboration and alkynylboration of alkenes, have been developed with broad scopes and a high tolerance to a variety of functional groups. Therefore, the Ni-catalyzed carboboration of alkenes has become an efficient synthetic protocol to deliver substituted compounds by the cross-coupling of alkenes, electrophiles, and B2Pin2. Despite this progress, a number of challenging issues remaining in the field include broadening the types of carboboration reactions, especially the asymmetric ones, diversifying electrophile types (which is limited to halogens for now) and gaining profound insight into the reaction mechanisms. This review summarizes the recent progress in this emerging field from the literature published since 2018. It will provide the scientific community with convenience to access collective information and to accelerate their further research in order to broaden the scope of methodology and application in drug discovery programs.

Novel hydroxyl carboximates derived from ß-elemene: design, synthesis and anti-tumour activities evaluation.

A series of novel N-alkyl-N-hydroxyl carboximates derived from ß-elemene were fortuitously discovered. Most of them showed more potent anti-proliferative activities than their lead compound ß-elemene (1). Notably, compound 11i exhibited significant inhibitory effects on the proliferation of three lung cell lines (H1975, A549 and H460) and several other tumour cell lines (H1299, U87MG, MV4-11, and SU-DHL-2). Preliminary mechanistic studies revealed that compound 11i could significantly induce cell apoptosis. Further in vivo study in the H460 xenograft mouse model validated the anti-tumour activities of 11i with a greater tumour growth inhibition (TGI, 68.3%) than ß-elemene and SAHA (50.1% and 55.9% respectively) at 60 mg/kg ip dosing, without obvious body weight loss and toxicity. Thus, such N-alkyl-N-hydroxyl carboximate class of compounds exemplified as 11i demonstrated potent anticancer activities both in vitro and in vivo, and should warrant further investigation for potential anticancer therapy.

Recent Progress in Fragmentation of Katritzky Salts Enabling Formation of C-C, C-B, and C-S Bonds.

Since their discovery in 1970s, Katritzky salts have emerged as one of the most important classes of building blocks for use in organic synthesis and drug discovery. These bulky pyridinium salts derived from alkylamine can readily generate alkyl radical and undergo a variety of organic transformation reactions such as alkylation, arylation, alkenylation, alkynylation, carbonylation, sulfonylation, and borylation. Through these transformations, complexed molecules bearing new C-C, C-B, or C-S bonds can be constructed in easy ways and in simple steps. This review aims to summarize recent advances in these versatile building blocks in well-classified categories. Representative examples and their reaction mechanisms are discussed. The hope is to provide the scientific community with convenient access to collective information and accelerate further research.

Recent Advances in Synergistic Antitumor Effects Exploited from the Inhibition of Ataxia Telangiectasia and RAD3-Related Protein Kinase (ATR).

As one of the key phosphatidylinositol 3-kinase-related kinases (PIKKs) family members, ataxia telangiectasia and RAD3-related protein kinase (ATR) is crucial in maintaining mammalian cell genomic integrity in DNA damage response (DDR) and repair pathways. Dysregulation of ATR has been found across different cancer types. In recent years, the inhibition of ATR has been proven to be effective in cancer therapy in preclinical and clinical studies. Importantly, tumor-specific alterations such as ATM loss and Cyclin E1 (CCNE1) amplification are more sensitive to ATR inhibition and are being exploited in synthetic lethality (SL) strategy. Besides SL, synergistic anticancer effects involving ATRi have been reported in an increasing number in recent years. This review focuses on the recent advances in different forms of synergistic antitumor effects, summarizes the pharmacological benefits and ongoing clinical trials behind the biological mechanism, and provides perspectives for future challenges and opportunities. The hope is to draw awareness to the community that targeting ATR should have great potential in developing effective anticancer medicines.

Protective Effects of Anthocyanins Extracted from Vaccinium Uliginosum on 661W Cells Against Microwave-Induced Retinal Damage / 中国结合医学杂志

OBJECTIVE@#To study the protective effect of anthocyanins extracted from Vaccinium Uliginosum (VU) on retinal 661W cells against microwave radiation induced retinal injury.@*METHODS@#661W cells were divided into 6 groups, including control, model [661W cells radiated by microwave (30 mW/cm2, 1 h)] and VU groups [661W cells pretreated with anthocyanins extracted from VU (25, 50, 100 and 200 µg/mL, respectively) for 48 h, and radiated by microwave 30 mW/cm2, 1 h]. After treatment with different interventions, the cell apoptosis index (AI) was determined using Heochst staining; contents of malonaldehyde (MDA), glutataione (GSH), and activity of superoxide dismutase (SOD) were measured. mRNA expressions of nuclear factor erythroid 2-related factor 2 (Nrf2) and heme oxygenase 1(HO-1) were detected by real time quantitative polymerase chain reaction, and the expression of HO-1 protein was examined by Western blot analysis. Nucleus and cytoplasm were separated and Nrf2 protein expression was further verified by Western blot analysis.@*RESULTS@#There was significant difference in AI among the groups (F=322.83, P<;0.05). Compared with the control group, AI was significantly higher in the model group and was lower in 4 VU-pretreated groups (P<;0.05). Linear regression analysis showed the decline of AI was in a dose-dependent manner with VU treatment (r=0.8419, P<;0.05). The MDA and GSH contents of 661W cells in VU-treated groups were significantly lower than the model group (P<;0.05). Compared with the model group, the SOD activity in the VU-treated groups (50, 100 and 200 µg/mL) was significantly higher (all P<;0.05). The Nrf2 and HO-1 mRNA expressions were slightly increased after irradiation, and obviously increased in 100 µg/mL VU-treated group. After irradiation, the relative expressions of HO-1 and Nrf2 proteins in nucleus were slightly increased (P<;0.05), and the changes in cytoplasm were not obvious, whereas it was significantly increased in both nucleus and cytoplasm in the VU treatment groups.@*CONCLUSIONS@#Anthocyanins extracted from VU could reduce apoptosis, stabilize cell membrane, and alleviate oxidant injury of mouse retinal photoreceptor 661W cells. The mechanism might be through activating Nrf2/HO-1 signal pathway and inducing HO-1 transcription and translation.

Protective Effects of Anthocyanins Extracted from Vaccinium Uliginosum on 661W Cells Against Microwave-Induced Retinal Damage.

OBJECTIVE: To study the protective effect of anthocyanins extracted from Vaccinium Uliginosum (VU) on retinal 661W cells against microwave radiation induced retinal injury. METHODS: 661W cells were divided into 6 groups, including control, model [661W cells radiated by microwave (30 mW/cm2, 1 h)] and VU groups [661W cells pretreated with anthocyanins extracted from VU (25, 50, 100 and 200 µg/mL, respectively) for 48 h, and radiated by microwave 30 mW/cm2, 1 h]. After treatment with different interventions, the cell apoptosis index (AI) was determined using Heochst staining; contents of malonaldehyde (MDA), glutataione (GSH), and activity of superoxide dismutase (SOD) were measured. mRNA expressions of nuclear factor erythroid 2-related factor 2 (Nrf2) and heme oxygenase 1(HO-1) were detected by real time quantitative polymerase chain reaction, and the expression of HO-1 protein was examined by Western blot analysis. Nucleus and cytoplasm were separated and Nrf2 protein expression was further verified by Western blot analysis. RESULTS: There was significant difference in AI among the groups (F=322.83, P<;0.05). Compared with the control group, AI was significantly higher in the model group and was lower in 4 VU-pretreated groups (P<;0.05). Linear regression analysis showed the decline of AI was in a dose-dependent manner with VU treatment (r=0.8419, P<;0.05). The MDA and GSH contents of 661W cells in VU-treated groups were significantly lower than the model group (P<;0.05). Compared with the model group, the SOD activity in the VU-treated groups (50, 100 and 200 µg/mL) was significantly higher (all P<;0.05). The Nrf2 and HO-1 mRNA expressions were slightly increased after irradiation, and obviously increased in 100 µg/mL VU-treated group. After irradiation, the relative expressions of HO-1 and Nrf2 proteins in nucleus were slightly increased (P<;0.05), and the changes in cytoplasm were not obvious, whereas it was significantly increased in both nucleus and cytoplasm in the VU treatment groups. CONCLUSIONS: Anthocyanins extracted from VU could reduce apoptosis, stabilize cell membrane, and alleviate oxidant injury of mouse retinal photoreceptor 661W cells. The mechanism might be through activating Nrf2/HO-1 signal pathway and inducing HO-1 transcription and translation.

Implementation and Application of an Intelligent Pterygium Diagnosis System Based on Deep Learning.

Objective: This study aims to implement and investigate the application of a special intelligent diagnostic system based on deep learning in the diagnosis of pterygium using anterior segment photographs. Methods: A total of 1,220 anterior segment photographs of normal eyes and pterygium patients were collected for training (using 750 images) and testing (using 470 images) to develop an intelligent pterygium diagnostic model. The images were classified into three categories by the experts and the intelligent pterygium diagnosis system: (i) the normal group, (ii) the observation group of pterygium, and (iii) the operation group of pterygium. The intelligent diagnostic results were compared with those of the expert diagnosis. Indicators including accuracy, sensitivity, specificity, kappa value, the area under the receiver operating characteristic curve (AUC), as well as 95% confidence interval (CI) and F1-score were evaluated. Results: The accuracy rate of the intelligent diagnosis system on the 470 testing photographs was 94.68%; the diagnostic consistency was high; the kappa values of the three groups were all above 85%. Additionally, the AUC values approached 100% in group 1 and 95% in the other two groups. The best results generated from the proposed system for sensitivity, specificity, and F1-scores were 100, 99.64, and 99.74% in group 1; 90.06, 97.32, and 92.49% in group 2; and 92.73, 95.56, and 89.47% in group 3, respectively. Conclusion: The intelligent pterygium diagnosis system based on deep learning can not only judge the presence of pterygium but also classify the severity of pterygium. This study is expected to provide a new screening tool for pterygium and benefit patients from areas lacking medical resources.

An online diabetic retinopathy screening tool for patients with type 2 diabetes.

AIM: To develop a useful diabetic retinopathy (DR) screening tool for patients with type 2 diabetes mellitus (T2DM). METHODS: A DR prediction model based on the Logistic regression algorithm was established on the development dataset containing 778 samples (randomly assigned to the training dataset and the internal validation dataset at a ratio of 7:3). The generalization capability of the model was assessed using an external validation dataset containing 128 samples. The DR risk calculator was developed through WeChat Developer Tools using JavaScript, which was embedded in the WeChat Mini Program. RESULTS: The model revealed risk factors (duration of diabetes, diabetic nephropathy, and creatinine level) and protective factors (annual DR screening and hyperlipidemia) for DR. In the internal and external validation, the recall ratios of the model were 0.92 and 0.89, respectively, and the area under the curve values were 0.82 and 0.70, respectively. CONCLUSION: The DR screening tool integrates education, risk prediction, and medical advice function, which could help clinicians in conducting DR risk assessments and providing recommendations for ophthalmic referral to increase the DR screening rate among patients with T2DM.

Attitudes of medical workers in China toward artificial intelligence in ophthalmology: a comparative survey.

BACKGROUND: In the development of artificial intelligence in ophthalmology, the ophthalmic AI-related recognition issues are prominent, but there is a lack of research into people's familiarity with and their attitudes toward ophthalmic AI. This survey aims to assess medical workers' and other professional technicians' familiarity with, attitudes toward, and concerns about AI in ophthalmology. METHODS: This is a cross-sectional study design study. An electronic questionnaire was designed through the app Questionnaire Star, and was sent to respondents through WeChat, China's version of Facebook or WhatsApp. The participation was voluntary and anonymous. The questionnaire consisted of four parts, namely the respondents' background, their basic understanding of AI, their attitudes toward AI, and their concerns about AI. A total of 562 respondents were counted, with 562 valid questionnaires returned. The results of the questionnaires are displayed in an Excel 2003 form. RESULTS: There were 291 medical workers and 271 other professional technicians completed the questionnaire. About 1/3 of the respondents understood AI and ophthalmic AI. The percentages of people who understood ophthalmic AI among medical workers and other professional technicians were about 42.6 % and 15.6 %, respectively. About 66.0 % of the respondents thought that AI in ophthalmology would partly replace doctors, about 59.07 % having a relatively high acceptance level of ophthalmic AI. Meanwhile, among those with AI in ophthalmology application experiences (30.6 %), above 70 % of respondents held a full acceptance attitude toward AI in ophthalmology. The respondents expressed medical ethics concerns about AI in ophthalmology. And among the respondents who understood AI in ophthalmology, almost all the people said that there was a need to increase the study of medical ethics issues in the ophthalmic AI field. CONCLUSIONS: The survey results revealed that the medical workers had a higher understanding level of AI in ophthalmology than other professional technicians, making it necessary to popularize ophthalmic AI education among other professional technicians. Most of the respondents did not have any experience in ophthalmic AI but generally had a relatively high acceptance level of AI in ophthalmology, and there was a need to strengthen research into medical ethics issues.

Five-Category Intelligent Auxiliary Diagnosis Model of Common Fundus Diseases Based on Fundus Images.

Purpose: The discrepancy of the number between ophthalmologists and patients in China is large. Retinal vein occlusion (RVO), high myopia, glaucoma, and diabetic retinopathy (DR) are common fundus diseases. Therefore, in this study, a five-category intelligent auxiliary diagnosis model for common fundus diseases is proposed; the model's area of focus is marked. Methods: A total of 2000 fundus images were collected; 3 different 5-category intelligent auxiliary diagnosis models for common fundus diseases were trained via different transfer learning and image preprocessing techniques. A total of 1134 fundus images were used for testing. The clinical diagnostic results were compared with the diagnostic results. The main evaluation indicators included sensitivity, specificity, F1-score, area under the concentration-time curve (AUC), 95% confidence interval (CI), kappa, and accuracy. The interpretation methods were used to obtain the model's area of focus in the fundus image. Results: The accuracy rates of the 3 intelligent auxiliary diagnosis models on the 1134 fundus images were all above 90%, the kappa values were all above 88%, the diagnosis consistency was good, and the AUC approached 0.90. For the 4 common fundus diseases, the best results of sensitivity, specificity, and F1-scores of the 3 models were 88.27%, 97.12%, and 84.02%; 89.94%, 99.52%, and 93.90%; 95.24%, 96.43%, and 85.11%; and 88.24%, 98.21%, and 89.55%, respectively. Conclusions: This study designed a five-category intelligent auxiliary diagnosis model for common fundus diseases. It can be used to obtain the diagnostic category of fundus images and the model's area of focus. Translational Relevance: This study will help the primary doctors to provide effective services to all ophthalmologic patients.

An Evaluation System of Fundus Photograph-Based Intelligent Diagnostic Technology for Diabetic Retinopathy and Applicability for Research.

INTRODUCTION: In April 2018, the US Food and Drug Administration (FDA) approved the world's first artificial intelligence (AI) medical device for detecting diabetic retinopathy (DR), the IDx-DR. However, there is a lack of evaluation systems for DR intelligent diagnostic technology. METHODS: Five hundred color fundus photographs of diabetic patients were selected. DR severity varied from grade 0 to 4, with 100 photographs for each grade. Following that, these were diagnosed by both ophthalmologists and the intelligent technology, the results of which were compared by applying the evaluation system. The system includes primary, intermediate, and advanced evaluations, of which the intermediate evaluation incorporated two methods. Main evaluation indicators were sensitivity, specificity, and kappa value. RESULTS: The AI technology diagnosed 93 photographs with no DR, 107 with mild non-proliferative DR (NPDR), 107 with moderate NPDR, 108 with severe NPDR, and 85 with proliferative DR (PDR). The sensitivity, specificity, and kappa value of the AI diagnoses in the primary evaluation were 98.8%, 88.0%, and 0.89, respectively. According to method 1 of the intermediate evaluation, the sensitivity of AI diagnosis was 98.0%, specificity 97.0%, and the kappa value 0.95. In method 2 of the intermediate evaluation, the sensitivity of AI diagnosis was 95.5%, the specificity 99.3%, and kappa value 0.95. In the advanced evaluation, the kappa value of the intelligent diagnosis was 0.86. CONCLUSIONS: This article proposes an evaluation system for color fundus photograph-based intelligent diagnostic technology of DR and demonstrates an application of this system in a clinical setting. The results from this evaluation system serve as the basis for the selection of scenarios in which DR intelligent diagnostic technology can be applied.