TM7SF2-induced lipid reprogramming promotes cell proliferation and migration via CPT1A/Wnt/ß-Catenin axis in cervical cancer cells.

Cervical cancer poses a serious threat to women's health globally. Our previous studies found that upregulation of TM7SF2, which works as an enzyme involved in the process of cholesterol biosynthesis expression, was highly correlated with cervical cancer. However, the mechanistic basis of TM7SF2 promoting cervical cancer progression via lipid metabolism remains poorly understood. Therefore, quantification of fatty acids and lipid droplets were performed in vitro and in vivo. The protein-protein interaction was verified by Co-IP technique. The mechanism and underlying signaling pathway of TM7SF2 via CPT1A associated lipid metabolism in cervical cancer development were explored using Western blotting, IHC, colony formation, transwell assay, and wound healing assay. This study reported that overexpression of TM7SF2 increased fatty acids content and lipid droplets both in vivo and in vitro experiments. While knockout of TM7SF2 obviously attenuated this process. Moreover, TM7SF2 directly bonded with CPT1A, a key enzyme in fatty acid oxidation, and regulated CPT1A protein expression in cervical cancer cells. Notably, the proliferation and metastasis of cervical cancer cells were elevated when their CPT1A expression was upregulated. Then, rescue assay identified that CPT1A overexpressed could enhance the cell viability and migration in TM7SF2-knockout cells. Furthermore, depletion of TM7SF2 significantly inhibited WNT and ß-catenin proteins expression, which was enhanced by CPT1A-overexpressed. The proliferation and migration of cervical cancer cells were reversed in CPT1A-overexpressed cells with the treatment of MSAB, an inhibitor of Wnt/ß-Catenin pathway. This study put forward an idea that TM7SF2-induced lipid reprogramming promotes proliferation and migration via CPT1A/Wnt/ß-Catenin axis in cervical cancer, underlying the progression of cervical cancer.

Developing deep learning-based strategies to predict the risk of hepatocellular carcinoma among patients with nonalcoholic fatty liver disease from electronic health records.

OBJECTIVE: The accuracy of deep learning models for many disease prediction problems is affected by time-varying covariates, rare incidence, covariate imbalance and delayed diagnosis when using structured electronic health records data. The situation is further exasperated when predicting the risk of one disease on condition of another disease, such as the hepatocellular carcinoma risk among patients with nonalcoholic fatty liver disease due to slow, chronic progression, the scarce of data with both disease conditions and the sex bias of the diseases. The goal of this study is to investigate the extent to which the aforementioned issues influence deep learning performance, and then devised strategies to tackle these challenges. These strategies were applied to improve hepatocellular carcinoma risk prediction among patients with nonalcoholic fatty liver disease. METHODS: We evaluated two representative deep learning models in the task of predicting the occurrence of hepatocellular carcinoma in a cohort of patients with nonalcoholic fatty liver disease (n = 220,838) from a national EHR database. The disease prediction task was carefully formulated as a classification problem while taking censorship and the length of follow-up into consideration. RESULTS: We developed a novel backward masking scheme to deal with the issue of delayed diagnosis which is very common in EHR data analysis and evaluate how the length of longitudinal information after the index date affects disease prediction. We observed that modeling time-varying covariates improved the performance of the algorithms and transfer learning mitigated reduced performance caused by the lack of data. In addition, covariate imbalance, such as sex bias in data impaired performance. Deep learning models trained on one sex and evaluated in the other sex showed reduced performance, indicating the importance of assessing covariate imbalance while preparing data for model training. CONCLUSIONS: The strategies developed in this work can significantly improve the performance of hepatocellular carcinoma risk prediction among patients with nonalcoholic fatty liver disease. Furthermore, our novel strategies can be generalized to apply to other disease risk predictions using structured electronic health records, especially for disease risks on condition of another disease.

Expression levels of ADAMTS 5, 9, and 12 in endometrial polyps and their predictive value for the diagnosis and recurrence of endometrial polyps.

PURPOSE: Endometrial polyps (EPs) are common gynecological disorders for which no clear etiology has been found. ADAMTS have been associated with a variety of diseases. This study aimed to investigate the potential correlation between serologic levels of ADAMTS 5, 9, and 12 in patients with EPs. METHODS: A total of 88 patients were categorized into two groups: the EPs group, consisting of recurrent EPs and first occurrence EPs, and a control group. The study compared the general information and serum levels of ADAMTS 5, 9, and 12 between the groups. RESULTS: Regarding the general data, a statistically significant age difference (p < 0.05) was observed, while no significant differences were found in the other variables. After considering age as a confounding factor, the previously observed statistical significance in the differences of ADAMTS5 and 9 between the groups diminished. However, it was found that the concentrations of ADAMTS12 in both the EPs group and the recurrent EPs group were significantly higher compared to the control group and the first occurrence EPs group (p < 0.05). ROC curves were generated to determine the critical values of ADAMTS12 for predicting EPs and recurrent EPs, which were found to be 0.6962 ng/ml (sensitivity: 100 %, specificity: 39.5 %) and 0.8768 ng/ml (sensitivity: 75.0 %, specificity: 76.3 %), respectively. CONCLUSION: Our findings revealed elevated serologic levels of ADAMTS12 in the EPs group, particularly in the recurrent EPs group. Furthermore, ADAMTS-12 was identified as a valuable biomarker for assisting in the diagnosis and prediction of EPs recurrence.

Histone lactylation inhibits RARγ expression in macrophages to promote colorectal tumorigenesis through activation of TRAF6-IL-6-STAT3 signaling.

Macrophages are phenotypically and functionally diverse in the tumor microenvironment (TME). However, how to remodel macrophages with a protumor phenotype and how to manipulate them for therapeutic purposes remain to be explored. Here, we show that in the TME, RARγ is downregulated in macrophages, and its expression correlates with poor prognosis in patients with colorectal cancer (CRC). In macrophages, RARγ interacts with tumor necrosis factor receptor-associated factor 6 (TRAF6), which prevents TRAF6 oligomerization and autoubiquitination, leading to inhibition of nuclear factor κB signaling. However, tumor-derived lactate fuels H3K18 lactylation to prohibit RARγ gene transcription in macrophages, consequently enhancing interleukin-6 (IL-6) levels in the TME and endowing macrophages with tumor-promoting functions via activation of signal transducer and activator of transcription 3 (STAT3) signaling in CRC cells. We identified that nordihydroguaiaretic acid (NDGA) exerts effective antitumor action by directly binding to RARγ to inhibit TRAF6-IL-6-STAT3 signaling. This study unravels lactate-driven macrophage function remodeling by inhibition of RARγ expression and highlights NDGA as a candidate compound for treating CRC.

Developing deep learning-based strategies to predict the risk of hepatocellular carcinoma among patients with nonalcoholic fatty liver disease from electronic health records.

Background: Deep learning models showed great success and potential when applied to many biomedical problems. However, the accuracy of deep learning models for many disease prediction problems is affected by time-varying covariates, rare incidence, and covariate imbalance when using structured electronic health records data. The situation is further exasperated when predicting the risk of one disease on condition of another disease, such as the hepatocellular carcinoma risk among patients with nonalcoholic fatty liver disease due to slow, chronic progression, the scarce of data with both disease conditions and the sex bias of the diseases. Objective: The goal of this study is to investigate the extent to which time-varying covariates, rare incidence, and covariate imbalance influence deep learning performance, and then devised strategies to tackle these challenges. These strategies were applied to improve hepatocellular carcinoma risk prediction among patients with nonalcoholic fatty liver disease. Methods: We evaluated two representative deep learning models in the task of predicting the occurrence of hepatocellular carcinoma in a cohort of patients with nonalcoholic fatty liver disease (n = 220,838) from a national EHR database. The disease prediction task was carefully formulated as a classification problem while taking censorship and the length of follow-up into consideration. Results: We developed a novel backward masking scheme to evaluate how the length of longitudinal information after the index date affects disease prediction. We observed that modeling time-varying covariates improved the performance of the algorithms and transfer learning mitigated reduced performance caused by the lack of data. In addition, covariate imbalance, such as sex bias in data impaired performance. Deep learning models trained on one sex and evaluated in the other sex showed reduced performance, indicating the importance of assessing covariate imbalance while preparing data for model training. Conclusions: Devising proper strategies to address challenges from time-varying covariates, lack of data, and covariate imbalance can be key to counteracting data bias and accurately predicting disease occurrence using deep learning models. The novel strategies developed in this work can significantly improve the performance of hepatocellular carcinoma risk prediction among patients with nonalcoholic fatty liver disease. Furthermore, our novel strategies can be generalized to apply to other disease risk predictions using structured electronic health records, especially for disease risks on condition of another disease.

Inhibition of SK Channels in VTA Affects Dopaminergic Neurons to Improve the Depression-Like Behaviors of Post-Stroke Depression Rats.

Purpose: This study aimed to investigate the effect of small-conductance calcium-activated potassium channels (SK channels) on the dopaminergic (DA) neuron pathways in the ventral tegmental area (VTA) during the pathogenesis of post-stroke depression (PSD) and explore the improvement of PSD by inhibiting the SK channels. Patients and Methods: Four groups of Sprague-Dawley rats were randomly divided: Control, PSD, SK channel inhibitor (apamin) and SK channel activator (CyPPA) groups. In both control and CyPPA groups, sham surgery was performed. In the other two groups, middle cerebral arteries were occluded. The behavioral indicators related to depression in different groups were compared. Immunofluorescence was used to measure the activity of DA neurons in the VTA, while qRT-PCR was used to assess the expression of SK channel genes. Results: The results showed that apamin treatment improved behavioral indicators related to depression compared to the PSD group. Furthermore, the qRT-PCR analysis revealed differential expression of the KCNN1 and KCNN3 subgenes of the SK channels in each group. Immunofluorescence analysis revealed an increase in the expression of DA neurons in the VTA of the PSD group, which was subsequently reduced upon apamin intervention. Conclusion: This study suggests that SK channel activation following stroke contributes to depression-related behaviors in PSD rats through increased expression of DA neurons in the VTA. And depression-related behavior is improved in PSD rats by inhibiting the SK channels. The results of this study provide a new understanding of PSD pathogenesis and the possibility of developing new strategies to prevent PSD by targeting SK channels.

Indications for Magnetic Resonance Imaging in Patients With Behcet Uveitis.

BACKGROUND: Behcet disease is a systemic vasculitis, which may involve the eyes and central nervous system. The true prevalence of neurological involvement is not precisely known but may be associated with ocular involvement. This study investigates the association between Behcet uveitis and neuro-Behcet disease. METHODS: A retrospective single-center analysis was conducted for consecutive patients with Behcet uveitis at the Massachusetts Eye Research and Surgery Institution. Uveitis characteristics, neurological symptoms, fluorescein fundus angiography, and MRI results were recorded. RESULTS: Our population included 108 patients with Behcet uveitis, and 26 (24.1%) were found to have neurological involvement associated with Behcet disease. Optic nerve leakage on fundus angiography and neurological symptoms were associated with an increased risk of neurological involvement. Three cases (11.5%) were nonparenchymal, while 23 (88.5%) were parenchymal with lesions in the cortex, subcortical white matter, thalamus, basal ganglia, and brainstem. CONCLUSIONS: There is a high comorbidity between ocular and neurological involvement in Behcet disease. Careful assessment of neurological symptoms and baseline fluorescein fundus angiography are recommended for patients with Behcet disease. MRI has a high diagnostic yield and should be pursued if there is concern for progressive or pre-existing neurological involvement.

SGReg: segmentation guided 3D/2D rigid registration for orthogonal X-ray and CT images in spine surgery navigation.

Objective.One of the essential technologies in various image-guided spine surgeries is the rigid registration of 3D pre-operative CT and 2D intra-operative X-ray images. The 3D/2D registration is patterned as two essential tasks, that is, dimensional correspondence establishment and estimation of the 3D pose. 3D data is projected to 2D for dimensional correspondence by most of the existing methods, which makes pose parameters difficult to estimate caused by the loss of spatial information. This work aims to develop a reconstruction based 3D/2D registration method for spine surgery navigation.Approach.A novel segmentation-guided 3D/2D registration (SGReg) method for orthogonal X-ray and CT images was proposed based on reconstruction. SGReg consists of a bi-path segmentation network and an inter-path multi-scale pose estimation module. The X-ray segmentation path in the bi-path segmentation network reconstructs 3D spatial information from 2D orthogonal X-ray images to segmentation masks; meanwhile, the CT segmentation path predicts segmentation masks from 3D CT images, thereby bringing the 3D/2D data into dimensional correspondence. In the inter-path multi-scale pose estimation module, the features from the two segmentation paths are integrated, and the pose parameters are directly regressed under the guidance of the coordinate information.Main result.We evaluated SGReg using a public dataset CTSpine1k and compared the registration performance with other methods. SGReg achieved considerable improvement over other methods with great robustness.SignificanceWe have proposed an end-to-end 3D/2D registration framework named SGReg. Based on the idea of reconstruction, SGReg performs a unified framework between dimensional correspondence establishment and direct pose estimation in 3D space, showing significant potential in spine surgery navigation.

Prediction of Brain Metastases Development in Patients With Lung Cancer by Explainable Artificial Intelligence From Electronic Health Records.

PURPOSE: Early detection of brain metastases (BMs) is critical for prompt treatment and optimal control of the disease. In this study, we seek to predict the risk of developing BM among patients diagnosed with lung cancer on the basis of electronic health record (EHR) data and to understand what factors are important for the model to predict BM development through explainable artificial intelligence approaches accurately. MATERIALS AND METHODS: We trained a recurrent neural network model, REverse Time AttentIoN (RETAIN), to predict the risk of developing BM using structured EHR data. To interpret the model's decision process, we analyzed the attention weights in the RETAIN model and the SHAP values from a feature attribution method, Kernel SHAP, to identify the factors contributing to BM prediction. RESULTS: We developed a high-quality cohort with 4,466 patients with BM from the Cerner Health Fact database, which contains over 70 million patients from more than 600 hospitals. RETAIN uses this data set to achieve the best area under the receiver operating characteristic curve at 0.825, a significant improvement over the baseline model. We also extended a feature attribution method, Kernel SHAP, to structured EHR data for model interpretation. Both RETAIN and Kernel SHAP can identify important features related to BM prediction. CONCLUSION: To the best of our knowledge, this is the first study to predict BM using structured EHR data. We achieved decent prediction performance for BM prediction and identified factors highly relevant to BM development. The sensitivity analysis demonstrated that both RETAIN and Kernel SHAP could discriminate unrelated features and put more weight on the features important to BM. Our study explored the potential of applying explainable artificial intelligence for future clinical applications.

Nicotinamide Adenine Dinucleotide Precursor Suppresses Hepatocellular Cancer Progression in Mice.

Targeting Nicotinamide adenine dinucleotide (NAD) metabolism has emerged as a promising anti-cancer strategy; we aimed to explore the health benefits of boosting NAD levels with nicotinamide riboside (NR) on hepatocellular carcinoma (HCC). We established three in vivo tumor models, including subcutaneous transplantation tumor model in both Balb/c nude mice (xenograft), C57BL/6J mice (allograft), and hematogenous metastatic neoplasm in nude mice. NR (400 mg/kg bw) was supplied daily in gavage. In-situ tumor growth or noninvasive bioluminescence were measured to evaluate the effect of NR on the HCC process. HepG2 cells were treated with transforming growth factor-ß (TGF-ß) in the absence/presence of NR in vitro. We found that NR supplementation alleviated malignancy-induced weight loss and metastasis to lung in nude mice in both subcutaneous xenograft and hematogenous metastasis models. NR supplementation decreased metastasis to the bone and liver in the hematogenous metastasis model. NR supplementation also significantly decreased the size of allografted tumors and extended the survival time in C57BL/6J mice. In vitro experiments showed that NR intervention inhibited the migration and invasion of HepG2 cells triggered by TGF-ß. In summary, our results supply evidence that boosting NAD levels by supplementing NR alleviates HCC progression and metastasis, which may serve as an effective treatment for the suppression of HCC progression.

Implantation of XEN gel stent in a patient with ocular cicatricial pemphigoid.

Purpose: Herein, we report a case of XEN gel stent implantation in a patient with ocular cicatricial pemphigoid that successfully reduced glaucoma topical medication at one year. Observations: A 76-year-old male patient presented with severe ocular cicatricial pemphigoid and advanced glaucoma who required several topical medications to control intraocular pressure. Despite successful reduction of ocular inflammation with immunomodulatory therapy, his topical medication regimen prevented total remission of ocular inflammation. One year after XEN gel stent implantation, his intraocular pressures were controlled without any topical medication, and he had no ocular inflammation off any immunomodulatory therapy. Conclusions and Importance: The XEN gel stent represents a useful intervention for glaucoma treatment even in the setting of severe ocular surface disease and can improve outcomes for concurrent inflammatory and glaucomatous pathology.

Developing a deep learning model to predict epilepsy recurrence in patients with focal cortical dysplasia type III.

Background: A sizable number of patients with focal cortical dysplasia (FCD) type III-related refractory epilepsy continue to experience seizures postsurgically. Deep learning models can automatically assess complex medical image characteristics and predict prognosis with higher efficiency. This study sought to determine whether T2-weighted fluid attenuated inversion recovery (T2W FLAIR) images could predict prognosis of FCD type III-related refractory epilepsy using a deep learning approach. Methods: Magnetic resonance imaging (MRI) images of 266 patients with FCD type III diagnosed between 2015 and 2019 were included in this retrospective analysis. A deep learning algorithm utilizing a convolutional neural network (CNN) was trained to classify T2W FLAIR images according to Engel's classification. The preprocessed original image and the region of interest (ROI) outlined by clinicians were input into our neural network separately and then together. Precision, sensitivity, specificity, receiver operating characteristic (ROC) curves, and areas under the ROC curves (AUCs) were computed as part of the statistical analyses of the network performance with varied inputs of the network model assessed. Results: The overall performance met the following metrics when the original image only was input: AUC of 96.22%, sensitivity of 84.47%, and specificity of 97.21%. The metrics were as follows when the ROI only was input: area under the ROC curve of 94.76%, sensitivity of 84.92%, and specificity of 96.24%. For the combined inputs, the metrics were as follows: AUC of 97.17%, sensitivity of 90.86%, and specificity of 96.63%. Conclusions: Deep learning used with conventional MRI can effectively predict the recurrence conditions of epilepsy. Artificial intelligence may help the design of clinical management and enable more precise and individualized prediction for postsurgical prognosis of FCD type III-related refractory epilepsy.

Overview of Corneal Transplantation for the Nonophthalmologist.

Corneal transplant is a procedure that aims to replace dysfunctional corneal tissue with a transparent graft and is one of the most widely performed transplant surgeries, but its public and professional awareness is low outside of ophthalmology. Corneal tissue consists of 5 major layers that serve to maintain its structural integrity and refractive shape: the epithelium, Bowman's layer, the stroma, Descemet's membrane, and the endothelium. Failure or irreversible damage to any layer of the cornea may be an indication for corneal transplant, and variants of this procedure may be full thickness or selectively lamellar. Complications related to corneal transplantation may occur anywhere from during surgery to years afterward, including rejection, dehiscence, cataract, and glaucoma. Complications should be managed by an ophthalmologist, but other physicians should be aware of prophylactic medications. Topical immunosuppressants and steroids are effective for preventing and treating rejection episodes, whereas there is little evidence to support the use of systemic immunosuppression. Eye protection is recommended for any corneal transplant recipient. Physicians should counsel patients on corneal donation, especially if outside the United States, where donor tissue is in short supply.

Vitamin D protects silica particles induced lung injury by promoting macrophage polarization in a KLF4-STAT6 manner.

Silicosis is one of the severest occupational diseases worldwide, manifesting as infiltration of inflammatory cells, excessive secretion of pro-inflammatory mediators and pulmonary diffuse fibrosis. Macrophages polarization to M2 is one of the major strategies that attenuates inflammatory response. Our previous study found that vitamin D could protect against silica-induced lung injury by damping the secretion of pro-inflammatory cytokines. Here we further identified that vitamin D attenuated silica particles-induced lung inflammation by regulating macrophage polarization in a KLF4-STAT6 manner. Myeloid-specific Stat6 knockout (cKO) mice were generated for in vivo studies. Primary macrophages purified from bronchoalveolar lavage fluid (BALF) of wildtype or Stat6 cKO mice and differentiated THP-1 cells were used for in vitro studies. Vitamin D was found to promote alveolar macrophage polarizing to M2 phenotype through the STAT6 signaling pathway, as demonstrated by worse lung inflammation and ablated protection of vitamin D in silica particles-instilled Stat6 cKO mice. Mechanismly, vitamin D upregulated KLF4 expression in the alveolar macrophage, which synergistically activated STAT6. Additionally, KLF4 was found to upregulate macrophages autophagy, which protected them from silica particles-induced oxidative stress and cell apoptosis. The protective effects of vitamin D were dismissed by silencing KLF4. Our study demonstrates the potential mechanism of vitamin D-mediated macrophage polarization and reveals the therapeutic application of vitamin D in inflammatory disease.

Open Globe Injuries: Review of Evaluation, Management, and Surgical Pearls.

Ocular trauma may either be closed globe or open globe. Open globe injuries are full-thickness defects of the eyewall and are often differentiated by the mechanisms of injury from which they are caused: sharp or blunt trauma. They are ocular emergencies and can lead to substantial visual morbidity. Without timely intervention, damage is irreversible and leads to permanent vision loss. The goals of evaluation are to identify the mechanism of injury, characterize the extent of injury, and gather relevant history. If an open globe is suspected, ophthalmologic consultation should be requested. Once an open globe is diagnosed, preparations for surgery should be made immediately and steps should be taken to avoid further injury. Intraocular infection risk is relatively high, requiring immediate empiric systemic antibiotics. Emergent surgical exploration and primary closure is indicated whenever possible. After initial closure, secondary surgery and revision may be needed to improve vision outcomes, followed by extensive follow-up. In this review, best practices for evaluation and management are reviewed, with particular focus on the surgical approach and techniques.

Aloe gel glucomannan induced colon cancer cell death via mitochondrial damage-driven PINK1/Parkin mitophagy pathway.

Mitophagy can selectively remove damaged mitochondria, which is critical in regulating mitochondrial homeostasis in diseases, such as cancer. Herein, we found that Aloe gel glucomannan (AGP) significantly inhibited the proliferation of colon cancer cells. RNA-seq analysis revealed that AGP upregulated autophagy, lysosome and mitochondrial fission signal pathways in colon cancer cell line CT26. Notably, AGP induced the accumulation of impaired and reactive oxygen species (ROS)-generating mitochondria, which triggered excessive mitophagy. Interestingly, the mitophagy activator enhanced AGP-induced mitophagy and cytotoxicity, whereas the mitophagy inhibitor reversed the influence of AGP. Furthermore, activation of PINK1/Parkin mitophagy pathway and transcription factor EB (TFEB) signaling was dependent on ROS overproduction. Taken together, these results indicated that AGP induced cytotoxic mitophagy through ROS-related PINK1/Parkin pathway and TFEB activation in CT26 cells. The research would provide theoretical basis for the development of AGP as a promising anticancer agent.

STAT6 inhibits ferroptosis and alleviates acute lung injury via regulating P53/SLC7A11 pathway.

Compelling evidences have revealed the emerging role of ferroptosis in the pathophysiological process of acute lung injury (ALI), but its modulation is not clear. Here, we identified that STAT6 acted as a critical regulator of epithelium ferroptosis during ALI. Firstly, STAT6 expression and activity were increased in the ALI mice models caused by crystalline silica (CS), LPS and X-ray exposure. Followed by confirming the contribution of ferroptosis in the above ALI with ferrostatin-1 and deferoxamine intervention, bioinformatic analyses revealed that STAT6 expression was negatively correlated with ferroptosis. Consistently, lung epithelium-specific depletion of STAT6 in mice or STAT6 knockdown in cultured epithelial cells exacerbated ferroptosis in the above ALI. While overexpression of STAT6 in lung epithelial cells attenuated the ferroptosis. Mechanistically, SLC7A11 is a typical ferroptosis-related gene and negatively regulated by P53. CREB-binding protein (CBP) is a critical acetyltransferase of P53 acetylation, showing valuable regulation on targets' transcription. Herein, we found that STAT6 negatively regulates ferroptosis through competitively binding with CBP, which inhibits P53 acetylation and transcriptionally restores SLC7A11 expression. Finally, pulmonary-specific STAT6 overexpression decreased the ferroptosis and attenuated CS and LPS induced lung injury. Our findings revealed that STAT6 is a pivotal regulator of ferroptosis, which may be a potential therapeutic target for the treatment of acute lung injury.

Ocular injury from plastic airsoft bullet through protective steel mesh mask.

The importance of protective eyewear during activities which involve high velocity projectiles is often emphasized to patients, however the material and design of such eyewear is also important. We present the case of a boy who sustained ocular injury from a plastic airsoft bullet while wearing a protective mask issued by the manufacturer. The patient was found to have decreased vision, a corneal abrasion and hyphema. The patient fully recovered with topical prednisolone, cyclopentolate, and moxifloxacin. It is important to advocate for using polycarbonate protective eye wear for our patients who are engaging in activities which involve high velocity projectiles.

Automatic segmentation model of intercondylar fossa based on deep learning: a novel and effective assessment method for the notch volume.

BACKGROUND: Notch volume is associated with anterior cruciate ligament (ACL) injury. Manual tracking of intercondylar notch on MR images is time-consuming and laborious. Deep learning has become a powerful tool for processing medical images. This study aims to develop an MRI segmentation model of intercondylar fossa based on deep learning to automatically measure notch volume, and explore its correlation with ACL injury. METHODS: The MRI data of 363 subjects (311 males and 52 females) with ACL injuries incurred during non-contact sports and 232 subjects (147 males and 85 females) with intact ACL were retrospectively analyzed. Each layer of intercondylar fossa was manually traced by radiologists on axial MR images. Notch volume was then calculated. We constructed an automatic segmentation system based on the architecture of Res-UNet for intercondylar fossa and used dice similarity coefficient (DSC) to compare the performance of segmentation systems by different networks. Unpaired t-test was performed to determine differences in notch volume between ACL-injured and intact groups, and between males and females. RESULTS: The DSCs of intercondylar fossa based on different networks were all more than 0.90, and Res-UNet showed the best performance. The notch volume was significantly lower in the ACL-injured group than in the control group (6.12 ± 1.34 cm3 vs. 6.95 ± 1.75 cm3, P < 0.001). Females had lower notch volume than males (5.41 ± 1.30 cm3 vs. 6.76 ± 1.51 cm3, P < 0.001). Males and females who had ACL injuries had smaller notch than those with intact ACL (p < 0.001 and p < 0.005). Men had larger notches than women, regardless of the ACL injuries (p < 0.001). CONCLUSION: Using a deep neural network to segment intercondylar fossa automatically provides a technical support for the clinical prediction and prevention of ACL injury and re-injury after surgery.