Applications of Raman spectroscopy in ocular biofluid detection.

Ophthalmic and many systemic diseases may damage the eyes, resulting in changes in the composition and content of biomolecules in ocular biofluids such as aqueous humor and tear. Therefore, the biomolecules in biofluids are potential biomarkers to reveal pathological processes and diagnose diseases. Raman spectroscopy is a non-invasive, label-free, and cost-effective technique to provide chemical bond information of biomolecules and shows great potential in the detection of ocular biofluids. This review demonstrates the applications of Raman spectroscopy technology in detecting biochemical components in aqueous humor and tear, then summarizes the current problems encountered for clinical applications of Raman spectroscopy and looks forward to possible approaches to overcome technical bottlenecks. This work may provide a reference for wider applications of Raman spectroscopy in biofluid detection and inspire new ideas for the diagnosis of diseases using ocular biofluids.

Prediction of Vestibular Schwannoma Surgical Outcome Using Deep Neural Network.

OBJECTIVE: To compare shallow machine learning models and deep neural network (DNN) model in prediction of vestibular schwannoma (VS) surgical outcome. METHODS: One hundred eighty-eight patients with VS were included; all underwent suboccipital retrosigmoid sinus approach, and preoperative magnetic resonance imaging recorded a series of patient characteristics. Degree of tumor resection was collected during surgery, and facial nerve function was evaluated on the eighth day after surgery. Potential predictors of VS surgical outcome were obtained by univariate analysis, including tumor diameter, tumor volume, tumor surface area, brain tissue edema, tumor property, and tumor shape. This study proposes a DNN framework to predict the prognosis of VS surgical outcomes based on potential predictors and compares it with a series of classic machine learning algorithms including logistic regression. RESULTS: The results showed that 3 predictors of tumor diameter, tumor volume, and tumor surface area were the most important prognostic factors for VS surgical outcomes, followed by tumor shape, while brain tissue edema and tumor property were the least influential. Different from shallow machine learning models, such as logistic regression with average performance (area under the curve: 0.8263; accuracy: 81.38%), the proposed DNN shows better performance, where area under the curve and accuracy were 0.8723 and 85.64%, respectively. CONCLUSIONS: Based on potential risk factors, DNN can be exploited to achieve preoperative automatic assessment of VS surgical outcomes, and its performance is significantly better than other methods. It is therefore highly warranted to continue to investigate their utility as complementary clinical tools in predicting surgical outcomes preoperatively.