Artificial Intelligence for the Determination of a Management Strategy for Diminutive Colorectal Polyps: Hype, Hope, or Help.

Most colorectal polyps are diminutive, and malignant potential for these polyps is uncommon, especially for those in the rectosigmoid. However, many diminutive polyps are still being resected to determine whether these are adenomas or serrated/hyperplastic polyps. Resecting all the diminutive polyps is not cost-effective. Therefore, gastroenterologists have proposed optical diagnosis using image-enhanced endoscopy for polyp characterization. These technologies have achieved favorable outcomes, but are not widely available. Artificial intelligence has been used in clinical medicine to classify lesions. Here, artificial intelligence technology for the characterization of colorectal polyps is discussed in a decision-making context regarding diminutive colorectal polyps.

Myeloid-Derived Suppressor Cells Mediate Inflammation Resolution in Humans and Mice with Autoimmune Uveoretinitis.

Resolution of inflammation is an active process that leads to tissue homeostasis and involves multiple cellular and molecular mechanisms. Myeloid-derived suppressor cells (MDSCs) have recently emerged as important cellular components in the resolution of inflammation because of their activities to suppress T cell activation. In this article, we show that HLA-DR-CD11b+CD33+CD14+ human MDSCs and CD11b+Ly6G-Ly6C+ mouse MDSCs markedly increased in patients and mice during and before the resolution phase of autoimmune uveoretinitis. CD11b+Ly6C+ monocytes isolated from autoimmune uveoretinitis mice were able to suppress T cell proliferation in culture, and adoptive transfer of the cells accelerated the remission of autoimmune uveoretinitis in mice. Alternatively, depletion of CD11b+Ly6C+ monocytes at the resolution phase, but not CD11b+Ly6G+ granulocytes, exacerbated the disease. These findings collectively indicate that monocytic MDSCs serve as regulatory cells mediating the resolution of autoimmune uveoretinitis.

Automated classification of gastric neoplasms in endoscopic images using a convolutional neural network.

BACKGROUND: Visual inspection, lesion detection, and differentiation between malignant and benign features are key aspects of an endoscopist's role. The use of machine learning for the recognition and differentiation of images has been increasingly adopted in clinical practice. This study aimed to establish convolutional neural network (CNN) models to automatically classify gastric neoplasms based on endoscopic images. METHODS: Endoscopic white-light images of pathologically confirmed gastric lesions were collected and classified into five categories: advanced gastric cancer, early gastric cancer, high grade dysplasia, low grade dysplasia, and non-neoplasm. Three pretrained CNN models were fine-tuned using a training dataset. The classifying performance of the models was evaluated using a test dataset and a prospective validation dataset. RESULTS: A total of 5017 images were collected from 1269 patients, among which 812 images from 212 patients were used as the test dataset. An additional 200 images from 200 patients were collected and used for prospective validation. For the five-category classification, the weighted average accuracy of the Inception-Resnet-v2 model reached 84.6 %. The mean area under the curve (AUC) of the model for differentiating gastric cancer and neoplasm was 0.877 and 0.927, respectively. In prospective validation, the Inception-Resnet-v2 model showed lower performance compared with the endoscopist with the best performance (five-category accuracy 76.4 % vs. 87.6 %; cancer 76.0 % vs. 97.5 %; neoplasm 73.5 % vs. 96.5 %; P  < 0.001). However, there was no statistical difference between the Inception-Resnet-v2 model and the endoscopist with the worst performance in the differentiation of gastric cancer (accuracy 76.0 % vs. 82.0 %) and neoplasm (AUC 0.776 vs. 0.865). CONCLUSION: The evaluated deep-learning models have the potential for clinical application in classifying gastric cancer or neoplasm on endoscopic white-light images.

FLUOCINOLONE ACETONIDE IMPLANT FOR VOGT-KOYANAGI-HARADA DISEASE: Three-Year Outcomes of Efficacy and Safety.

PURPOSE: To investigate the efficacy and safety of fluocinolone acetonide intravitreal implant in patients with Vogt-Koyanagi-Harada disease. METHODS: A post hoc, subgroup analysis on patients with Vogt-Koyanagi-Harada was performed using data sets from two multicenter randomized trials on fluocinolone acetonide implant. Each subject received fluocinolone acetonide implantation in one eye and standard-of-care treatment in the fellow eye and was followed for 3 years. RESULTS: Thirty patients were included with the mean age of 38.5 years. The cumulative rate of uveitis recurrence for 3 years was significantly reduced in implanted eyes compared with fellow eyes (33 vs. 87%; P < 0.001). The reduction of daily corticosteroid dose was well maintained (12.8 mg before implantation vs. 3.7 mg after implantation; P = 0.001), but final vision was similar to preoperative vision in the implanted eyes (P = 0.082) and in the fellow eyes (P = 0.187). Postoperative elevation of intraocular pressure was more frequent in the implanted eyes than in the fellow eyes (70 vs. 20%; P < 0.001). Cataract progression occurred in all phakic implanted eyes. CONCLUSION: Fluocinolone acetonide intravitreal implant reduced uveitis recurrence rate and the dosage of systemic corticosteroid and immunosuppressant requirement in patients with Vogt-Koyanagi-Harada. However, cataract and intraocular pressure elevation developed frequently.

Complications of Pathologic Myopia.

Pathologic myopia (PM) is one of the leading causes of visual impairment worldwide. The pathophysiology of PM is not fully understood, but the axial elongation of the eye followed by chorioretinal thinning is suggested as a key mechanism. Pathologic myopia may lead to many complications such as chorioretinal atrophy, foveoschisis, choroidal neovascularization, rhegmatogenous retinal detachment, cataract, and glaucoma. Some complications affect visual acuity significantly, showing poor visual prognosis. This article aims to review the types, pathophysiology, treatment, and visual outcome of the complications of PM.

Risk factors for intraocular involvement in patients with primary central nervous system lymphoma.

To determine the risk factors for intraocular involvement in patients with primary central nervous system lymphoma (PCNSL), a retrospective chart review was performed on 136 patients who were pathologically diagnosed with PCNSL. The patients were investigated for demographics, clinical manifestation, and the profile of immunohistochemical tumor biomarkers, as well as for the presence of intraocular involvement of lymphoma at diagnosis or during follow-up. The mean age of the entire cohort was 58.6 ± 12.4 years, and the mean follow-up period was 31.1 ± 30.8 months. Twenty-nine (21 %) patients had an intraocular involvement, among which 20 (69 %) patients presented with intraocular involvement at diagnosis of PCNSL and 9 (31 %) patients developed intraocular involvement after a mean period of 32.4 ± 33.6 months. Of the patients with intraocular involvement, 8 (28 %) had no visual symptom at the diagnosis of ocular invasion. Between those with and without intraocular involvement, no significant differences were found with respect to the age, sex, and follow-up period as well as cerebrospinal fluid spread and bone marrow involvement. Among the immunohistochemical biomarkers, the Ki-67 proliferation index was significantly higher in patients with intraocular involvement than in patients without (P = 0.021), but the other investigated biomarkers did not show a significant difference between the two groups. A Ki-67 level ≥80 % was a risk factor for the intraocular involvement in patients with PCNSL (odds ratio, 2.63). Median overall survival was 39.0 months in the entire cohort and was not significantly different between those with and without intraocular involvement (P = 0.959).

Surgical outcomes according to vitreous management after scleral fixation of posterior chamber intraocular lenses.

PURPOSE: To investigate the effect of vitreous management on surgical outcomes after scleral fixation of posterior chamber intraocular lenses. METHODS: The medical records of 83 eyes from 83 patients, who underwent scleral fixation of posterior chamber intraocular lenses without any preexisting vitreoretinal complication and were followed up for ≥ 6 months, were reviewed retrospectively. Subjects were divided according to the strategy used for vitreous management into a pars plana vitrectomy (PPV) group (47 eyes) and an anterior vitrectomy (AV) group (36 eyes). Surgical outcomes were compared between the groups. RESULTS: Both groups were comparable regarding demographics, follow-up period, and underlying ocular disease. The degree of visual improvement was similar in both groups (P = 0.911). Postoperatively, each group developed a myopic shift, which was greater in the PPV group than in the AV group (P = 0.040). Intraocular pressure elevations ≥ 25 mmHg occurred in 50% of eyes in the AV group and in 11% of eyes in the PPV group (P < 0.001). Intraocular lenses dislocated more frequently in the AV group (28%) than in the PPV group (9%; P = 0.036). Intraocular lens capture was more common in the PPV group (23%) than in the AV group (3%; P = 0.010). There was no significant difference in the rate of postoperative vitreoretinal complications. CONCLUSION: In eyes that received scleral fixation of posterior chamber intraocular lenses, PPV decreased the chances of postoperative intraocular pressure elevation and intraocular lens dislocation but increased the likelihood of intraocular lens capture and the magnitude of postoperative myopic shift.

Performance reporting design in artificial intelligence studies using image-based TNM staging and prognostic parameters in rectal cancer: a systematic review.

PURPOSE: The integration of artificial intelligence (AI) and magnetic resonance imaging in rectal cancer has the potential to enhance diagnostic accuracy by identifying subtle patterns and aiding tumor delineation and lymph node assessment. According to our systematic review focusing on convolutional neural networks, AI-driven tumor staging and the prediction of treatment response facilitate tailored treat-ment strategies for patients with rectal cancer. METHODS: This paper summarizes the current landscape of AI in the imaging field of rectal cancer, emphasizing the performance reporting design based on the quality of the dataset, model performance, and external validation. RESULTS: AI-driven tumor segmentation has demonstrated promising results using various convolutional neural network models. AI-based predictions of staging and treatment response have exhibited potential as auxiliary tools for personalized treatment strategies. Some studies have indicated superior performance than conventional models in predicting microsatellite instability and KRAS status, offer-ing noninvasive and cost-effective alternatives for identifying genetic mutations. CONCLUSION: Image-based AI studies for rectal can-cer have shown acceptable diagnostic performance but face several challenges, including limited dataset sizes with standardized data, the need for multicenter studies, and the absence of oncologic relevance and external validation for clinical implantation. Overcoming these pitfalls and hurdles is essential for the feasible integration of AI models in clinical settings for rectal cancer, warranting further research.

Deep Learning Prediction of Axillary Lymph Node Metastasis in Breast Cancer Patients Using Clinical Implication-Applied Preprocessed CT Images.

Background: Accurate detection of axillary lymph node (ALN) metastases in breast cancer is crucial for clinical staging and treatment planning. This study aims to develop a deep learning model using clinical implication-applied preprocessed computed tomography (CT) images to enhance the prediction of ALN metastasis in breast cancer patients. Methods: A total of 1128 axial CT images of ALN (538 malignant and 590 benign lymph nodes) were collected from 523 breast cancer patients who underwent preoperative CT scans between January 2012 and July 2022 at Hallym University Medical Center. To develop an optimal deep learning model for distinguishing metastatic ALN from benign ALN, a CT image preprocessing protocol with clinical implications and two different cropping methods (fixed size crop [FSC] method and adjustable square crop [ASC] method) were employed. The images were analyzed using three different convolutional neural network (CNN) architectures (ResNet, DenseNet, and EfficientNet). Ensemble methods involving and combining the selection of the two best-performing CNN architectures from each cropping method were applied to generate the final result. Results: For the two different cropping methods, DenseNet consistently outperformed ResNet and EfficientNet. The area under the receiver operating characteristic curve (AUROC) for DenseNet, using the FSC and ASC methods, was 0.934 and 0.939, respectively. The ensemble model, which combines the performance of the DenseNet121 architecture for both cropping methods, delivered outstanding results with an AUROC of 0.968, an accuracy of 0.938, a sensitivity of 0.980, and a specificity of 0.903. Furthermore, distinct trends observed in gradient-weighted class activation mapping images with the two cropping methods suggest that our deep learning model not only evaluates the lymph node itself, but also distinguishes subtler changes in lymph node margin and adjacent soft tissue, which often elude human interpretation. Conclusions: This research demonstrates the promising performance of a deep learning model in accurately detecting malignant ALNs in breast cancer patients using CT images. The integration of clinical considerations into image processing and the utilization of ensemble methods further improved diagnostic precision.

Comparison of outcomes between modified double-flanged sutureless scleral fixation and conventional sutured scleral fixation.

This retrospective study aimed to compare the outcomes of modified double-flanged sutureless scleral fixation versus sutured scleral fixation. Medical records of 65 eyes from 65 patients who underwent double-flanged scleral fixation (flange group) or conventional scleral fixation (suture group) between 2021 and 2022 were reviewed. Visual and refractive outcomes, as well as postoperative complications, were compared 1, 2, and 6 months after surgery. We included 31 eyes in the flange group and 34 eyes in the suture group. At 6 months postoperatively, the flange group showed better uncorrected visual acuity (0.251 ± 0.328 vs. 0.418 ± 0.339 logMAR, P = 0.041) and a smaller myopic shift (- 0.74 ± 0.93 vs. - 1.33 ± 1.15 diopter, P = 0.007) compared to the suture group. The flange group did not experience any instances of iris capture, while the suture group had iris capture in 10 eyes (29.4%; P < 0.001). In the flange group, all intraocular lenses remained centered, whereas in the suture group, they were decentered in 8 eyes (23.5%; P = 0.005). The double-flanged technique not only prevented iris capture and decentration of the intraocular lens but also reduced myopic shift by enhancing the stability of the intraocular lens.

Estimating infant age from skull X-ray images using deep learning.

This study constructed deep learning models using plain skull radiograph images to predict the accurate postnatal age of infants under 12 months. Utilizing the results of the trained deep learning models, it aimed to evaluate the feasibility of employing major changes visible in skull X-ray images for assessing postnatal cranial development through gradient-weighted class activation mapping. We developed DenseNet-121 and EfficientNet-v2-M convolutional neural network models to analyze 4933 skull X-ray images collected from 1343 infants. Notably, allowing for a ± 1 month error margin, DenseNet-121 reached a maximum corrected accuracy of 79.4% for anteroposterior (AP) views (average: 78.0 ± 1.5%) and 84.2% for lateral views (average: 81.1 ± 2.9%). EfficientNet-v2-M reached a maximum corrected accuracy 79.1% for AP views (average: 77.0 ± 2.3%) and 87.3% for lateral views (average: 85.1 ± 2.5%). Saliency maps identified critical discriminative areas in skull radiographs, including the coronal, sagittal, and metopic sutures in AP skull X-ray images, and the lambdoid suture and cortical bone density in lateral images, marking them as indicators for evaluating cranial development. These findings highlight the precision of deep learning in estimating infant age through non-invasive methods, offering the progress for clinical diagnostics and developmental assessment tools.

Topographic correlation between ß-zone parapapillary atrophy and retinal nerve fiber layer defect.

OBJECTIVE: To investigate whether a topographic correlation exists between ß-zone parapapillary atrophy (PPA) and retinal nerve fiber layer (RNFL) defect. The location and extent of the ß-zone were examined. DESIGN: Retrospective, cross-sectional study. PARTICIPANTS: One hundred twenty-eight eyes from 128 consecutive patients with primary open-angle glaucoma (POAG) and a single localized RNFL defect were included. METHODS: Digital optic disc photographs of the enrolled eyes were reviewed and eyes with ß-zone PPA were identified. The topographic parameters of ß-zone PPA and RNFL defect were measured on optic disc photographs and digital red-free RNFL photographs. The association between these parameters was examined statistically. MAIN OUTCOME MEASURES: Angular location and angular extent of ß-zone PPA and RNFL defect, angular location of point of maximum radial extent (PMRE) of ß-zone PPA, and ß-zone PPA-to-disc area ratio. RESULTS: Eighty-two (64.1%) of the 128 eyes with a single localized RNFL defect had ß-zone PPA. Patients with ß-zone PPA were younger (by 6.6 years) than those without ß-zone PPA (P = 0.001). ß-Zone PPA was located most commonly inferotemporally (65.9%). The RNFL defect was located in the same hemifield as the ß-zone PPA in 76% of eyes and was located in the same hemifield as PMRE in 88% of eyes. The angular location of the RNFL defect showed a linear correlation with those of ß-zone PPA (r = 0.390; P<0.001) and PMRE (r = 0.558; P<0.001). The angular extent of RNFL defect was not correlated significantly with that of ß-zone PPA (P = 0.106), but it was associated weakly with ß-zone PPA-to-disc area ratio (r = 0.197; P = 0.026). The angular extent of the RNFL defect also was correlated with the cup-to-disc ratio (r = 0.322; P<0.001) and the cup-to-disc area ratio (r = 0.337; P = 0.002). CONCLUSIONS: In POAG, a localized RNFL defect is correlated spatially with ß-zone PPA.

Association between subconjunctival hemorrhage and hemorrhagic disorders: a nationwide population-based study.

Subconjunctival hemorrhage (SCH) is a benign eye condition that is often noticeable and leads to medical attention. Despite previous studies investigating the relationship between SCH and cardiovascular diseases, the relationship between SCH and bleeding disorders remains controversial. In order to gain further insight into this association, a nationwide cohort study was conducted using data from the National Health Insurance Service-National Sample Cohort version 2.0 from 2006 to 2015. The study defined SCH using a diagnostic code and compared the incidence and risk factors of intracerebral hemorrhage (ICH) and gastrointestinal (GI) bleeding in 36,772 SCH individuals and 147,088 propensity score (PS)-matched controls without SCH. The results showed that SCH was associated with a lower risk of ICH (HR = 0.76, 95% CI = 0.622-0.894, p = 0.002) and GI bleeding (HR = 0.816, 95% CI = 0.690-0.965, p = 0.018) when compared to the PS-matched control group. This reduced risk was more pronounced in females and in the older age group (≥ 50 years), but not observed in males or younger age groups. In conclusion, SCH dose not increase the risk of ICH and major GI bleeding and is associated with a decreased incidence in females and individuals aged ≥ 50 years.

Clinical Features and Risk Factors of Uveitis in Korean Children with Juvenile Idiopathic Arthritis: A Retrospective Cohort Study.

This study aimed to investigate the clinical features and risk factors of uveitis in Korean children with juvenile idiopathic arthritis (JIA). The medical records of JIA patients diagnosed between 2006 and 2019 and followed up for ≥1 year were retrospectively reviewed, and various factors including laboratory findings were analyzed for the risk of developing uveitis. JIA-associated uveitis (JIA-U) developed in 30 (9.8%) of 306 JIA patients. The mean age at the first uveitis development was 12.4 ± 5.7 years, which was 5.6 ± 3.7 years after the JIA diagnosis. The common JIA subtypes in the uveitis group were oligoarthritis-persistent (33.3%) and enthesitis-related arthritis (30.0%). The uveitis group had more baseline knee joint involvement (76.7% vs. 51.4%), which increased the risk of JIA-U during follow-up (p = 0.008). Patients with the oligoarthritis-persistent subtype developed JIA-U more frequently than those without it (20.0% vs. 7.8%; p = 0.016). The final visual acuity of JIA-U was tolerable (0.041 ± 0.103 logMAR). In Korean children with JIA, JIA-U may be associated with the oligoarthritis-persistent subtype and knee joint involvement.

Differentiating malignant and benign eyelid lesions using deep learning.

Artificial intelligence as a screening tool for eyelid lesions will be helpful for early diagnosis of eyelid malignancies and proper decision-making. This study aimed to evaluate the performance of a deep learning model in differentiating eyelid lesions using clinical eyelid photographs in comparison with human ophthalmologists. We included 4954 photographs from 928 patients in this retrospective cross-sectional study. Images were classified into three categories: malignant lesion, benign lesion, and no lesion. Two pre-trained convolutional neural network (CNN) models, DenseNet-161 and EfficientNetV2-M architectures, were fine-tuned to classify images into three or two (malignant versus benign) categories. For a ternary classification, the mean diagnostic accuracies of the CNNs were 82.1% and 83.0% using DenseNet-161 and EfficientNetV2-M, respectively, which were inferior to those of the nine clinicians (87.0-89.5%). For the binary classification, the mean accuracies were 87.5% and 92.5% using DenseNet-161 and EfficientNetV2-M models, which was similar to that of the clinicians (85.8-90.0%). The mean AUC of the two CNN models was 0.908 and 0.950, respectively. Gradient-weighted class activation map successfully highlighted the eyelid tumors on clinical photographs. Deep learning models showed a promising performance in discriminating malignant versus benign eyelid lesions on clinical photographs, reaching the level of human observers.

Development and Application of an Active Pharmacovigilance Framework Based on Electronic Healthcare Records from Multiple Centers in Korea.

INTRODUCTION: With the availability of retrospective pharmacovigilance data, the common data model (CDM) has been identified as an efficient approach towards anonymized multicenter analysis; however, the establishment of a suitable model for individual medical systems and applications supporting their analysis is a challenge. OBJECTIVE: The aim of this study was to construct a specialized Korean CDM (K-CDM) for pharmacovigilance systems based on a clinical scenario to detect adverse drug reactions (ADRs). METHODS: De-identified patient records (n = 5,402,129) from 13 institutions were converted to the K-CDM. From 2005 to 2017, 37,698,535 visits, 39,910,849 conditions, 259,594,727 drug exposures, and 30,176,929 procedures were recorded. The K-CDM, which comprises three layers, is compatible with existing models and is potentially adaptable to extended clinical research. Local codes for electronic medical records (EMRs), including diagnosis, drug prescriptions, and procedures, were mapped using standard vocabulary. Distributed queries based on clinical scenarios were developed and applied to K-CDM through decentralized or distributed networks. RESULTS: Meta-analysis of drug relative risk ratios from ten institutions revealed that non-steroidal anti-inflammatory drugs (NSAIDs) increased the risk of gastrointestinal hemorrhage by twofold compared with aspirin, and non-vitamin K anticoagulants decreased cerebrovascular bleeding risk by 0.18-fold compared with warfarin. CONCLUSION: These results are similar to those from previous studies and are conducive for new research, thereby demonstrating the feasibility of K-CDM for pharmacovigilance. However, the low quality of original EMR data, incomplete mapping, and heterogeneity between institutions reduced the validity of the analysis, thus necessitating continuous calibration among researchers, clinicians, and the government.

The Incidence and Risk Factors of Medial and Inferior Orbital Wall Fractures in Korea: A Nationwide Cohort Study.

PURPOSE: We aimed to investigate orbital wall fracture incidence and risk factors in the general Korean population. METHOD: The Korea National Health Insurance Service-National Sample Cohort dataset was analyzed to find subjects with an orbital wall fracture between 2011 and 2015 (based on the diagnosis code) and to identify incident cases involving a preceding disease-free period of 8 years. The incidence of orbital wall fracture in the general population was estimated, and the type of orbital wall fracture was categorized. Sociodemographic risk factors were also examined using Cox regression analysis. RESULTS: Among 1,080,309 cohort subjects, 2415 individuals with newly diagnosed orbital wall fractures were identified. The overall incidence of orbital wall fractures was estimated as 46.19 (95% CI: 44.37-48.06) per 100,000 person-years. The incidence was high at 10-29 and 80+ years old and showed a male predominance with an average male-to-female ratio of 3.33. The most common type was isolated inferior orbital wall fracture (59.4%), followed by isolated medial orbital wall fracture (23.7%), combination fracture (15.0%), and naso-orbito-ethmoid fracture (1.5%). Of the fracture patients, 648 subjects (26.8%) underwent orbital wall fracture repair surgeries. Male sex, rural residence, and low income were associated with an increased risk of orbital wall fractures. CONCLUSIONS: The incidence of orbital wall fractures in Korea varied according to age groups and was positively associated with male sex, rural residency, and low economic income. The most common fracture type was an isolated inferior orbital wall fracture.

Automated histological classification for digital pathology images of colonoscopy specimen via deep learning.

Colonoscopy is an effective tool to detect colorectal lesions and needs the support of pathological diagnosis. This study aimed to develop and validate deep learning models that automatically classify digital pathology images of colon lesions obtained from colonoscopy-related specimen. Histopathological slides of colonoscopic biopsy or resection specimens were collected and grouped into six classes by disease category: adenocarcinoma, tubular adenoma (TA), traditional serrated adenoma (TSA), sessile serrated adenoma (SSA), hyperplastic polyp (HP), and non-specific lesions. Digital photographs were taken of each pathological slide to fine-tune two pre-trained convolutional neural networks, and the model performances were evaluated. A total of 1865 images were included from 703 patients, of which 10% were used as a test dataset. For six-class classification, the mean diagnostic accuracy was 97.3% (95% confidence interval [CI], 96.0-98.6%) by DenseNet-161 and 95.9% (95% CI 94.1-97.7%) by EfficientNet-B7. The per-class area under the receiver operating characteristic curve (AUC) was highest for adenocarcinoma (1.000; 95% CI 0.999-1.000) by DenseNet-161 and TSA (1.000; 95% CI 1.000-1.000) by EfficientNet-B7. The lowest per-class AUCs were still excellent: 0.991 (95% CI 0.983-0.999) for HP by DenseNet-161 and 0.995 for SSA (95% CI 0.992-0.998) by EfficientNet-B7. Deep learning models achieved excellent performances for discriminating adenocarcinoma from non-adenocarcinoma lesions with an AUC of 0.995 or 0.998. The pathognomonic area for each class was appropriately highlighted in digital images by saliency map, particularly focusing epithelial lesions. Deep learning models might be a useful tool to help the diagnosis for pathologic slides of colonoscopy-related specimens.

Prediction of White Matter Hyperintensity in Brain MRI Using Fundus Photographs via Deep Learning.

PURPOSE: We investigated whether a deep learning algorithm applied to retinal fundoscopic images could predict cerebral white matter hyperintensity (WMH), as represented by a modified Fazekas scale (FS), on brain magnetic resonance imaging (MRI). METHODS: Participants who had undergone brain MRI and health-screening fundus photography at Hallym University Sacred Heart Hospital between 2010 and 2020 were consecutively included. The subjects were divided based on the presence of WMH, then classified into three groups according to the FS grade (0 vs. 1 vs. 2+) using age matching. Two pre-trained convolutional neural networks were fine-tuned and evaluated for prediction performance using 10-fold cross-validation. RESULTS: A total of 3726 fundus photographs from 1892 subjects were included, of which 905 fundus photographs from 462 subjects were included in the age-matched balanced dataset. In predicting the presence of WMH, the mean area under the receiver operating characteristic curve was 0.736 ± 0.030 for DenseNet-201 and 0.724 ± 0.026 for EfficientNet-B7. For the prediction of FS grade, the mean accuracies reached 41.4 ± 5.7% with DenseNet-201 and 39.6 ± 5.6% with EfficientNet-B7. The deep learning models focused on the macula and retinal vasculature to detect an FS of 2+. CONCLUSIONS: Cerebral WMH might be partially predicted by non-invasive fundus photography via deep learning, which may suggest an eye-brain association.