Gradient Boosted Decision Tree Classification of Endophthalmitis Versus Uveitis and Lymphoma from Aqueous and Vitreous IL-6 and IL-10 Levels

Establishment of a prediction tool for ocular trauma patients with machine learning algorithm

AIM

To predict final visual acuity and analyze significant factors influencing open globe injury prognosis.

METHODS

Prediction models were built using a supervised classification algorithm from Microsoft Azure Machine Learning Studio. The best algorithm was selected to analyze the predicted final visual acuity. We retrospectively reviewed the data of 171 patients with open globe injury who visited the Pusan National University Hospital between January 2010 and July 2020. We then applied cross-validation, the permutation feature importance method, and the synthetic minority over-sampling technique to enhance tool performance.

RESULTS

The two-class boosted decision tree model showed the best predictive performance. The accuracy, precision, recall, F1 score, and area under the receiver operating characteristic curve were 0.925, 0.962, 0.833, 0.893, and 0.971, respectively. To increase the efficiency and efficacy of the prognostic tool, the top 14 features were finally selected using the permutation feature importance method: (listed in the order of importance) retinal detachment, location of laceration, initial visual acuity, iris damage, surgeon, past history, size of the scleral laceration, vitreous hemorrhage, trauma characteristics, age, corneal injury, primary diagnosis, wound location, and lid laceration.

CONCLUSION

Here we devise a highly accurate model to predict the final visual acuity of patients with open globe injury. This tool is useful and easily accessible to doctors and patients, reducing the socioeconomic burden. With further multicenter verification using larger datasets and external validation, we expect this model to become useful worldwide.

TMJOAI: An Artificial Web-Based Intelligence Tool for Early Diagnosis of the Temporomandibular Joint Osteoarthritis

Osteoarthritis is a chronic disease that affects the temporomandibular joint (TMJ), causing chronic pain and disability. To diagnose patients suffering from this disease before advanced degradation of the bone, we developed a diagnostic tool called TMJOAI. This machine learning based algorithm is capable of classifying the health status TMJ in of patients using 52 clinical, biological and jaw condyle radiomic markers. The TMJOAI includes three parts. the feature preparation, selection and model evaluation. Feature generation includes the choice of radiomic features (condylar trabecular bone or mandibular fossa), the histogram matching of the images prior to the extraction of the radiomic markers, the generation of feature pairwise interaction, etc.; the feature selection are based on the p-values or AUCs of single features using the training data; the model evaluation compares multiple machine learning algorithms (e.g. regression-based, tree-based and boosting algorithms) from 10 times 5-fold cross validation. The best performance was achieved with averaging the predictions of XGBoost and LightGBM models; and the inclusion of 32 additional markers from the mandibular fossa of the joint improved the AUC prediction performance from 0.83 to 0.88. After cross-validation and testing, the tools presented here have been deployed on an open-source, web-based system, making it accessible to clinicians. TMJOAI allows users to add data and automatically train and update the machine learning models, and therefore improve their performance.

Cytological diagnosis of vitreoretinal lymphomas: A case series

OBJECTIVE

To assess the cytological diagnosis and follow-up of patients suffering from vitreoretinal lymphoma (VRL) diagnosed in our institution.

METHODS AND RESULTS

From January 2010 to June 2017, we collected 15 patients with VRL. Twelve patients had diffuse large B-cell lymphoma (DLBCL); of these, 11 had primary central nervous system (CNS) DLBCL, one had ocular localisation of follicular lymphoma, one had extranodal NK/T-cell nasal type lymphoma and one had chronic lymphocytic leukaemia. The results of the cytological examination (cell morphology and immunocytochemistry) of the vitreous fluid were available for 9/15 VRL. The interleukin-10/-6 ratio was >1 in eight of 12 DLBCL. Molecular testing was useful in 6/15 cases (clonality evaluation or MYD88 L265P mutation testing). Eight out of 11 primary CNS DLBCL patients had CNS involvement, with 22-month progression-free survival. In our series, only two out of 11 CNS DLBCL patients died of disease after 2 and 5 years, respectively.

CONCLUSIONS

The short delay to assert the diagnosis of VRL could explain the quite good prognosis in our series, which highlights the need to consider a diagnosis of DLBCL as first step. The cytological features, as a reliable way to identify VRL, must always guide the choice of techniques for further investigations given the small amount of vitreous fluid available for analysis.

Regulatory Role of rno-miR-30b-5p in IL-10 and Toll-like Receptor 4 Expressions of T Lymphocytes in Experimental Autoimmune Uveitis In Vitro

Uveitis is a serious eye disease that usually damages young adult's health. MicroRNAs (miRNAs) are a class of small noncoding RNAs which regulate messenger RNA (mRNA) expression. It is predicted that rno-miR-30b-5p can regulate the expressions of interleukin-10 (IL-10) and Toll-like receptor 4 (TLR4). In this study, the regulatory role of rno-miR-30b-5p in IL-10 and TLR4 gene expressions was validated using luciferase activity assay. Further, the inflammatory manifestation of the anterior segment and pathological examination of the eye were explored in experimental autoimmune uveitis (EAU) rats. Meanwhile, the levels of rno-miR-30b-5p in eye tissues, spleen, and lymph nodes were measured using quantitative PCR (Q-PCR). IL-10 and TLR4 in spleen and lymph nodes were further separately determined by using Q-PCR and Enzyme-Linked Immunosorbent Assay (ELISA). Moreover, rno-miR-30b-5p mimic and its inhibitor were separately transfected into purified T cells, and the levels of IL-10 and TLR4 were detected using PCR, flow cytometry, and ELISA techniques. Results indicate that rno-miR-30b-5p was downregulated in spleen, lymph nodes, and eye tissues whereas the expressions of IL-10 and TLR4 at mRNA and protein levels were upregulated. The levels of IL-10 and TLR4 were negatively correlated to rno-miR-30b-5p levels. The result of in vitro cell transfection experiment indicates that IL-10 and TLR4 expressions were inhibited at mRNA and protein levels after T cells incubated with rno-miR-30b-5p mimic. However, the IL-10 and TLR4 mRNA levels were upregulated in purified T cells from spleen and lymph nodes after treatment with miR-30b-5p antagonist. In addition, there was no evident change of IL-10 and TLR4 proteins in spleen and lymph node T cells between EAU control and negative treatment groups. Flow cytometry analysis revealed that rno-miR-30b-5p mimic could reduce the number of both IL-10 and TLR4 positive cells, whereas rno-miR-30b-5p inhibitor could increase the number of IL-10 and TLR4 positive cells. Our study demonstrates that rno-miR-30b-5p influences the development of uveitis by regulating the level of IL-10 and TLR4 positive cells, thereby playing a role in the pathogenesis of uveitis.

Cytokines in uveitis

PURPOSE OF REVIEW

Increasing evidence supports Th17 cells as key mediators of ocular inflammatory disease. Cytokines that are important for the development and pathologic function of these cells are potential therapeutic targets in patients with immune mediated uveitis. This review provides an overview of these cytokines including recent insights about their roles in ocular inflammation from laboratory and clinical studies.

RECENT FINDINGS

Interleukin (IL)-6, IL-10, IL-17, IL-22, IL-23 and tumour necrosis factor-alpha (TNFα) are cytokines that have been examined for their functional role in uveitis and their relationship to pathologic Th17 cells. Studies in animal models, particularly in experimental autoimmune uveitis (EAU), have been instrumental in studying the role of these cytokines in disease pathogenesis. More recently, studies on aqueous, vitreous and serum from patients with uveitis using flow cytometry and multiplex ELISA bead-based methodologies have provided insights into the contribution of Th17 cells and the related cytokines in ocular inflammatory diseases. The central role of IL-23 in determining the pathologic Th17 fate has made it an effective therapeutic target in systemic diseases such as psoriasis and thereby an attractive potential target for patients with immune-mediated uveitis.

SUMMARY

Th17 cells, and their related cytokines, are important inflammatory mediators in autoimmune uveitis. Animal and human studies continue to provide new information to direct development of new cytokine-targeted therapies for patients with uveitis.