Evaluation of Methods for Detection and Semantic Segmentation of the Anterior Capsulotomy in Cataract Surgery Video.

Background: The capsulorhexis is one of the most important and challenging maneuvers in cataract surgery. Automated analysis of the anterior capsulotomy could aid surgical training through the provision of objective feedback and guidance to trainees. Purpose: To develop and evaluate a deep learning-based system for the automated identification and semantic segmentation of the anterior capsulotomy in cataract surgery video. Methods: In this study, we established a BigCat-Capsulotomy dataset comprising 1556 video frames extracted from 190 recorded cataract surgery videos for developing and validating the capsulotomy recognition system. The proposed system involves three primary stages: video preprocessing, capsulotomy video frame classification, and capsulotomy segmentation. To thoroughly evaluate its efficacy, we examined the performance of a total of eight deep learning-based classification models and eleven segmentation models, assessing both accuracy and time consumption. Furthermore, we delved into the factors influencing system performance by deploying it across various surgical phases. Results: The ResNet-152 model employed in the classification step of the proposed capsulotomy recognition system attained strong performance with an overall Dice coefficient of 92.21%. Similarly, the UNet model with the DenseNet-169 backbone emerged as the most effective segmentation model among those investigated, achieving an overall Dice coefficient of 92.12%. Moreover, the time consumption of the system was low at 103.37 milliseconds per frame, facilitating its application in real-time scenarios. Phase-wise analysis indicated that the Phacoemulsification phase (nuclear disassembly) was the most challenging to segment (Dice coefficient of 86.02%). Conclusion: The experimental results showed that the proposed system is highly effective in intraoperative capsulotomy recognition during cataract surgery and demonstrates both high accuracy and real-time capabilities. This system holds significant potential for applications in surgical performance analysis, education, and intraoperative guidance systems.

Adaptive Tensor-Based Feature Extraction for Pupil Segmentation in Cataract Surgery.

Cataract surgery remains the only definitive treatment for visually significant cataracts, which are a major cause of preventable blindness worldwide. Successful performance of cataract surgery relies on stable dilation of the pupil. Automated pupil segmentation from surgical videos can assist surgeons in detecting risk factors for pupillary instability prior to the development of surgical complications. However, surgical illumination variations, surgical instrument obstruction, and lens material hydration during cataract surgery can limit pupil segmentation accuracy. To address these problems, we propose a novel method named adaptive wavelet tensor feature extraction (AWTFE). AWTFE is designed to enhance the accuracy of deep learning-powered pupil recognition systems. First, we represent the correlations among spatial information, color channels, and wavelet subbands by constructing a third-order tensor. We then utilize higher-order singular value decomposition to eliminate redundant information adaptively and estimate pupil feature information. We evaluated the proposed method by conducting experiments with state-of-the-art deep learning segmentation models on our BigCat dataset consisting of 5,700 annotated intraoperative images from 190 cataract surgeries and a public CaDIS dataset. The experimental results reveal that the AWTFE method effectively identifies features relevant to the pupil region and improved the overall performance of segmentation models by up to 2.26% (BigCat) and 3.31% (CaDIS). Incorporation of the AWTFE method led to statistically significant improvements in segmentation performance (P < 1.29 × 10-10 for each model) and yielded the highest-performing model overall (Dice coefficients of 94.74% and 96.71% for the BigCat and CaDIS datasets, respectively). In performance comparisons, the AWTFE consistently outperformed other feature extraction methods in enhancing model performance. In addition, the proposed AWTFE method significantly improved pupil recognition performance by up to 2.87% in particularly challenging phases of cataract surgery.

CatStep: Automated Cataract Surgical Phase Classification and Boundary Segmentation Leveraging Inflated 3D-Convolutional Neural Network Architectures and BigCat.

Objective: Accurate identification of surgical phases during cataract surgery is essential for improving surgical feedback and performance analysis. Time spent in each surgical phase is an indicator of performance, and segmenting out specific phases for further analysis can simplify providing both qualitative and quantitative feedback on surgical maneuvers. Study Design: Retrospective surgical video analysis. Subjects: One hundred ninety cataract surgical videos from the BigCat dataset (comprising nearly 4 million frames, each labeled with 1 of 11 nonoverlapping surgical phases). Methods: Four machine learning architectures were developed for segmentation of surgical phases. Models were trained using cataract surgical videos from the BigCat dataset. Main Outcome Measures: Models were evaluated using metrics applied to frame-by-frame output and, uniquely in this work, metrics applied to phase output. Results: The final model, CatStep, a combination of a temporally sensitive model (Inflated 3D Densenet) and a spatially sensitive model (Densenet169), achieved an F1-score of 0.91 and area under the receiver operating characteristic curve of 0.95. Phase-level metrics showed considerable boundary segmentation performance with a median absolute error of phase start and end time of just 0.3 seconds and 0.1 seconds, respectively, a segmental F1-score @70 of 0.94, an oversegmentation score of 0.89, and a segmental edit score of 0.92. Conclusion: This study demonstrates the feasibility of high-performance automated surgical phase identification for cataract surgery and highlights the potential for improved surgical feedback and performance analysis. Financial Disclosures: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Tensor-based Feature Extraction for Pupil Recognition in Cataract Surgery.

Cataract surgery remains the definitive treatment for cataracts, which are a major cause of preventable blindness worldwide. Adequate and stable dilation of the pupil are necessary for the successful performance of cataract surgery. Pupillary instability is a known risk factor for cataract surgery complications, and the accurate segmentation of the pupil from surgical video streams can enable the analysis of intraoperative pupil changes in cataract surgery. However, pupil segmentation performance can suffer due to variations in surgical illumination, obscuration of the pupil with surgical instruments, and hydration of the lens material intraoperatively. To overcome these challenges, we present a novel method called tensor-based pupil feature extraction (TPFE) to improve the accuracy of pupil recognition systems. We analyzed the efficacy of this approach with experiments performed on a dataset of 4,560 intraoperative annotated images from 190 cataract surgeries in human patients. Our results indicate that TPFE can identify features relevant to pupil segmentation and that pupil segmentation with state-of-the-art deep learning models can be significantly improved with the TPFE method.

Optical Biometry Changes Throughout Childhood and Adolescence in Patients Wearing Ortho-K Lenses.

Background: Orthokeratology has been shown to suppress progressive myopia in some children. We examine the changes in optical biometry parameters in orthokeratology (Ortho-K) patients, in a retrospective longitudinal study at a tertiary eye care center in Ann Arbor, MI, USA. Methods: Optical biometry measurements obtained with the Lenstar LS 900 (Haag-Streit USA Inc, EyeSuite software version i9.1.0.0) were aggregated from 170 patients who had undergone Ortho-K for myopia correction between 5 and 20 years of age. Pre-intervention biometry measurements were compared with follow-up measurements done 6-18 months after initiation of Ortho-K. Linear mixed models were used to quantify associations in biometry changes with age of intervention allowing for correlation between measurements on two eyes of the same patient. Results: A total of 91 patients were included in the study. Axial length increased through the age of 15.7 ± 0.84 years for Ortho-K patients at our center. The growth curve in our Ortho-K population was comparable to previously published normal growth curves in Wuhan and Germany populations. Corneal thickness and keratometry decreased at a stable rate regardless of age of intervention (-7.9 µm, 95% CI [-10.2, -5.7], p < 0.001). Conclusion: In our population, Ortho-K did not appear to affect the overall trajectory of axial length progression when compared to normal growth curves, despite showing a previously described reduction in corneal thickness. As Ortho-K has been shown to have varying effects that differ from individual to individual, it continues to be important to reassess its effects on new populations to better understand its ideal uses.

Machine Learning Model Identifies Preoperative Opioid Use, Male Sex, and Elevated Body Mass Index as Predictive Factors for Prolonged Opioid Consumption Following Arthroscopic Meniscal Surgery.

PURPOSE: To develop a predictive machine learning model to identify prognostic factors for continued opioid prescriptions after arthroscopic meniscus surgery. METHODS: Patients undergoing arthroscopic meniscal surgery, such as meniscus debridement, repair, or revision at a single institution from 2013 to 2017 were retrospectively followed up to 1 year postoperatively. Procedural details were recorded, including concomitant procedures, primary versus revision, and whether a partial debridement or a repair was performed. Intraoperative arthritis severity was measured using the Outerbridge Classification. The number of opioid prescriptions in each month was recorded. Primary analysis used was the multivariate Cox-Regression model. We then created a naïve Bayesian model, a machine learning classifier that uses Bayes' theorem with an assumption of independence between variables. RESULTS: A total of 581 patients were reviewed. Postoperative opioid refills occurred in 98 patients (16.9%). Multivariate logistic modeling was used; independent risk factors for opioid refills included male sex, larger body mass index, and chronic preoperative opioid use, while meniscus resection demonstrated decreased likelihood of refills. Concomitant procedures, revision procedures, and presence of arthritis graded by the Outerbridge classification were not significant predictors of postoperative opioid refills. The naïve Bayesian model for extended postoperative opioid use demonstrated good fit with our cohort with an area under the curve of 0.79, sensitivity of 94.5%, positive predictive value (PPV) of 83%, and a detection rate of 78.2%. The two most important features in the model were preoperative opioid use and male sex. CONCLUSION: After arthroscopic meniscus surgery, preoperative opioid consumption and male sex were the most significant predictors for sustained opioid use beyond 1 month postoperatively. Intraoperative arthritis was not an independent risk factor for continued refills. A machine learning algorithm performed with high accuracy, although with a high false positive rate, to function as a screening tool to identify patients filling additional narcotic prescriptions after surgery. LEVEL OF EVIDENCE: III, retrospective comparative study.

Hemodialysis Patients Have Impaired Cerebrovascular Reactivity to CO2 Compared to Chronic Kidney Disease Patients and Healthy Controls: A Pilot Study.

INTRODUCTION: Recurrent hemodialysis (HD)-induced ischemia has emerged as a mechanism responsible for cognitive impairment in HD patients. Impairment of cerebrovascular function in HD patients may render the brain vulnerable to HD-induced ischemic injury. Cerebrovascular reactivity to CO2 (CVR) is a noninvasive marker of cerebrovascular function. Whether CVR is impaired in HD patients is unknown. In this study, we compared CVR between healthy participants, HD patients, and chronic kidney disease (CKD) patients not yet requiring dialysis. METHODS: This was a single-center prospective observational study carried out at Kidney Clinical Research Unit in London, Canada. We used carefully controlled hypercapnia to interrogate brain vasomotor control. Transcranial Doppler was combined with 10-mm Hg step changes in CO2 from baseline to hypercapnia (intervention) and back to baseline (recovery) to assess CVR in 8 HD, 10 CKD, and 17 heathy participants. RESULTS: HD patients had lower CVR than CKD or healthy participants during both intervention and recovery (P < 0.0001). There were no differences in CVR between healthy and CKD participants during either intervention (P = 0.88) or recovery (P = 0.99). The impaired CVR in HD patients was independent of CO2-induced changes in blood pressure, heart rate, cardiac output, or dialysis vintage. In the CKD group, CVR was not associated with the estimated glomerular filtration rate. CONCLUSIONS: Our study shows that HD patients have impaired CVR relative to CKD and healthy participants. This renders HD patients vulnerable to ischemic injury during circulatory stress of dialysis and may contribute to the pathogenesis of cognitive impairment.

Two-step machine learning method for the rapid analysis of microvascular flow in intravital video microscopy.

Microvascular blood flow is crucial for tissue and organ function and is often severely affected by diseases. Therefore, investigating the microvasculature under different pathological circumstances is essential to understand the role of the microcirculation in health and sickness. Microvascular blood flow is generally investigated with Intravital Video Microscopy (IVM), and the captured images are stored on a computer for later off-line analysis. The analysis of these images is a manual and challenging process, evaluating experiments very time consuming and susceptible to human error. Since more advanced digital cameras are used in IVM, the experimental data volume will also increase significantly. This study presents a new two-step image processing algorithm that uses a trained Convolutional Neural Network (CNN) to functionally analyze IVM microscopic images without the need for manual analysis. While the first step uses a modified vessel segmentation algorithm to extract the location of vessel-like structures, the second step uses a 3D-CNN to assess whether the vessel-like structures have blood flowing in it or not. We demonstrate that our two-step algorithm can efficiently analyze IVM image data with high accuracy (83%). To our knowledge, this is the first application of machine learning for the functional analysis of microvascular blood flow in vivo.

Cerebral Blood Flow Deviations in Critically Ill Patients: Potential Insult Contributing to Ischemic and Hyperemic Injury.

Background: Ischemic and hyperemic injury have emerged as biologic mechanisms that contribute to cognitive impairment in critically ill patients. Spontaneous deviations in cerebral blood flow (CBF) beyond ischemic and hyperemic thresholds may represent an insult that contributes to this brain injury, especially if they accumulate over time and coincide with impaired autoregulation. Methods: We used transcranial Doppler to measure the proportion of time that CBF velocity (CBFv) deviated beyond previously reported ischemic and hyperemic thresholds in a cohort of critically ill patients with respiratory failure and/or shock within 48 h of ICU admission. We also assessed whether these CBFv deviations were more common during periods of impaired dynamic autoregulation, and whether they are explained by concurrent variations in mean arterial pressure (MAP) and end-tidal PCO2 (PetCO2). Results: We enrolled 12 consecutive patients (three females) who were monitored for a mean duration of 462.6 ± 39.8 min. Across patients, CBFv deviated by more than 20-30% from its baseline for 17-24% of the analysis time. These CBFv deviations occurred equally during periods of preserved and impaired autoregulation, while concurrent variations in MAP and PetCO2 explained only 13-21% of these CBFv deviations. Conclusion: CBFv deviations beyond ischemic and hyperemic thresholds are common in critically ill patients with respiratory failure or shock. These deviations occur irrespective of the state of dynamic autoregulation and are not explained by changes in MAP and CO2. Future studies should explore mechanisms responsible for these CBFv deviations and establish whether their cumulative burden predicts poor neurocognitive outcomes.