Search Results (35)

Background: Occult lymph node metastasis (LNM) occurs in 30–80% of patients with clinically node-negative papillary thyroid carcinoma (cN0-PTC), partly owing to the limited sensitivity of current preoperative nodal assessment, and may contribute to postoperative recurrence. Conventional sentinel lymph node (SLN) biopsy, typically performed with a single tracer, has limited reliability for detecting occult metastatic nodes, which can result in either overtreatment or undertreatment with lymph node dissection. We aimed to develop a highly accurate multimodal prediction framework to accurately identify second-echelon lymph node metastasis (SeLNM) and non-sentinel lymph node metastasis (NsLNM). Methods: We prospectively enrolled 301 patients with cN0-PTC between April and October 2024, of whom 131 met the inclusion criteria. Intraoperatively, a dual-tracer technique combining carbon nanoparticles and indocyanine green was applied, and near-infrared imaging was used to record the entire SLN visualization process in real time. For each case, a 3 min video clip (150 frames) was captured. Two senior surgeons delineated regions of interest to generate 19,650 mask images. A total of 2048 spatial features and 20 temporal features were extracted, combined with 32 clinical variables, including demographics, ultrasound characteristics, and gene mutation status. Nine deep learning models were developed and evaluated using 10-fold cross-validation. Model performance was quantified using receiver operating characteristic curves, decision curve analysis curves, calibration curves, precision–recall curves, learning curves, and 12 metrics. Statistical comparisons were performed using the DeLong test, and models were further evaluated using a probability-based ranking approach. Shapley Additive Explanations (SHAP) analysis was applied to interpret key predictive features. The primary outcomes were SeLNM and NsLNM, defined based on postoperative histopathology. Results: The Long Short-Term Memory (LSTM) + Transformer model showed the best performance for both prediction tasks, with stable AUCs across training and testing (SeLNM: 0.980/0.982; NsLNM: 0.986/0.983). In the testing set, the model reached the same accuracy for both outcomes (94.7%) and showed strong sensitivity/specificity for SeLNM (94.7%/94.6%) and NsLNM (96.4%/91.5%). SHAP analysis indicated that time-series fluorescence flow features were the most influential predictors, followed by spatial structural features and SLN status. Conclusions: Dual-tracer SLN mapping with deep learning demonstrated encouraging intraoperative prediction of lymph node metastasis with interpretable features in this single-center cohort. Independent multicenter validation and prospective outcome studies are needed before considering clinical adoption. Full article

Introduction: Artificial intelligence (AI) and robot-assisted platforms are increasingly influencing cardiothoracic surgery. AI enhances risk prediction, imaging interpretation, and early complication detection, while robotics improves visualization, dexterity, and minimally invasive access. This systematic review evaluates the current evidence supporting these technologies and their implications for clinical practice. Methods: A systematic literature search was conducted across PubMed, Embase, Scopus, Web of Science, and Google Scholar (January 2000–May 2025) following PRISMA 2020 guidelines. After screening and eligibility assessment, 67 studies met predefined inclusion criteria and were incorporated into the qualitative synthesis. Additional high-impact reviews and consensus documents were consulted for contextual interpretation. Results: Machine learning models demonstrated modest but consistent improvements in predictive performance compared with EuroSCORE II and STS scores, particularly in high-risk cohorts. Robot-assisted mitral and coronary procedures showed reduced postoperative pain, blood loss, ICU stay, and recovery time in experienced centers, though early learning phases were associated with longer operative, cross-clamp, and bypass times. AI-enabled intraoperative tools, such as video analysis, workflow recognition, and real-time anatomical segmentation, emerged as promising adjuncts for surgical precision. Structured robotic training programs, especially simulation-based and dual-console pathways, accelerated proficiency acquisition. Conclusions: AI and robotic systems act as augmentative technologies that enhance rather than replace the surgeon’s role. Their safe and effective adoption requires standardized training, transparent AI decision pathways, and clear ethical and medico-legal governance. Full article

Objectives: To evaluate the short-term safety (30-day and in-hospital morbidity and mortality) and technical feasibility of uniportal video-assisted thoracic surgery (U-VATS) for anatomical lung resection in octogenarians (≥80 years) compared with younger patients (<80 years) at a single center. Methods: Ninety consecutive patients undergoing U-VATS anatomical lung resections between January 2020 and January 2024 were retrospectively analyzed. Patients were stratified by age: 60 patients < 80 years and 30 octogenarians ≥ 80 years. Propensity score matching (nearest-neighbor, 1:2 ratio, caliper 0.2 SD) yielded a matched cohort of 60 patients (40 younger, 20 octogenarians) for comparative analysis. Results: After matching, standardized mean differences (SMD) were <0.25 for most covariates, indicating good balance. Octogenarians demonstrated lower FEV1 (75.2 ± 15.3% vs. 87.5 ± 18.2%, p = 0.012) and DLCO (68.4 ± 12.1% vs. 78.5 ± 14.3%, p = 0.009), consistent with age-related pulmonary changes. Charlson Comorbidity Index was higher (5.3 ± 1.2 vs. 3.8 ± 1.4, p = 0.001). Surgical parameters were comparable: operative time (143.80 ± 42.3 vs. 136.55 ± 38.7 min, p = 0.524), blood loss (median 80 [IQR 50–120] vs. 95 [IQR 60–130] mL, p = 0.742). Zero conversions occurred. Major complications (Clavien–Dindo ≥ 3) occurred in 10% vs. 0% (absolute risk difference 10%, 95% CI: −3.2% to 23.2%). No 30-day mortality. 90-day mortality: 5% vs. 0% (p = 0.333); one-year: 15% vs. 0% (p = 0.035). Conclusions: U-VATS is technically feasible in carefully selected octogenarians with comparable intraoperative parameters to younger patients. Postoperative recovery differed meaningfully, with higher delirium rates, longer hospitalization, and greater rehabilitation needs. One-year mortality was higher in octogenarians, reflecting competing comorbid risk rather than surgical harm. Residual imbalance in comorbidity burden and pulmonary reserve after matching limits causal inference. These hypothesis-generating findings support U-VATS in selected octogenarians when comprehensive geriatric assessment and structured delirium prevention guide perioperative management; validation in larger multicenter prospective studies is required. Full article

Background/Objectives: Selection of the optimal recipient artery in superficial temporal artery to middle cerebral artery (STA–MCA) extracranial–intracranial bypass surgery is essential to ensure adequate cerebral perfusion. Various pre- and intraoperative tools for target vessel selection have been proposed. Indocyanine green fluorescence video angiography (ICG-VA) enables real-time visualization of cerebral hemodynamics, facilitating recipient vessel selection and anastomotic evaluation. Here, we review the literature and present the use of qualitative ICG-VA to support intraoperative decision-making during double-barrel (DB) STA–MCA bypass surgery. Case description: We report the case of a 68-year-old patient with bilateral steno-occlusive cerebrovascular disease, who developed progressive hemodynamic compromise of the left hemisphere after prior right-sided STA-MCA bypass. Preoperative imaging demonstrated impaired perfusion and posterior-to-anterior leptomeningeal collateralization from the posterior cerebral artery. During the left-sided DB bypass surgery, intravenous ICG-VA was used to assess relative cortical perfusion. Two superficial M4 branches with the most pronounced perfusion delay were selected as recipients based on the ICG-VA and anatomical criteria. Postoperative angiography confirmed graft patency. At short-term follow-up, the patient remained neurologically stable, with complete regression of preoperative symptoms. Conclusions: This case illustrates the application of qualitative ICG-VA for perfusion-oriented recipient vessel selection in DB STA-MCA bypass for steno-occlusive disease. Real-time perfusion assessment may complement conventional anatomical criteria for recipient vessel selection in flow-augmentation procedures. Further studies incorporating quantitative hemodynamic analysis are warranted. Full article

Background: Esophagectomy remains a technically demanding oncologic procedure with substantial morbidity, despite ongoing advances in minimally invasive and robotic techniques. Limitations in intraoperative visualization and anatomical recognition contribute to complications such as nerve injury and bleeding. Artificial intelligence (AI)-based intraoperative video analysis has emerged as a potential adjunct to enhance surgical perception and safety, but its application in esophagectomy has not been comprehensively reviewed. Methods: A systematic review was conducted in accordance with PRISMA guidelines. PubMed, Scopus, and Web of Science were searched without a lower date limit to identify eligible studies published up to January 2026, capturing early and contemporary applications of intraoperative AI in esophagectomy. Human studies involving any surgical approach were included. Data on the AI task, methodology, validation strategy, performance metrics, and reported clinical outcomes was extracted. Risk of bias was assessed using the ROBINS-I tool. Results: Six studies met the inclusion criteria, predominantly evaluating AI-driven analysis of intraoperative video during minimally invasive or robotic esophagectomy. Reported applications included real-time anatomical structure recognition, recurrent laryngeal nerve segmentation, detection of excessive nerve traction, instrument and event recognition, and surgical phase identification. Across studies, AI systems demonstrated performance comparable to expert surgeons for selected tasks and achieved real-time or near–real-time inference. One study reported earlier detection of excessive recurrent laryngeal nerve traction compared to conventional nerve integrity monitoring. However, most studies were retrospective, single-center, and feasibility-focused, with limited external validation and minimal assessment of patient-centered clinical outcomes. Conclusions: Artificial intelligence-based intraoperative analysis in esophagectomy is increasingly achievable and may enhance anatomical recognition, intraoperative risk detection, and procedural awareness. Nevertheless, current evidence remains preliminary, heterogeneous, and largely exploratory. Prospective, multicenter studies with standardized reporting and clinically meaningful outcome evaluation are required before routine implementation. Until such data is available, AI should be regarded as a complementary intraoperative tool rather than a standalone clinical decision-making system. Full article

Background: This single-center, retrospective, non-randomized observational study aims to explore the outcomes of video-assisted thoracoscopic surgery (VATS) wedge resection using the traditional clamp-and-suture technique versus staplers, with a focus on cost-effectiveness, operative time, and short-term postoperative outcomes. Methods: Data from 59 patients who underwent VATS wedge resection between 2018 and 2024 were retrospectively analyzed. Patients were divided into the stapler group (S-group, n = 27) and the clamp-and-suture group (C-group, n = 32). Technique selection was made intraoperatively by the surgeon based on lesion characteristics. Co-primary outcomes were total hospitalization cost and air leak duration > 2 days. Secondary outcomes included drainage time, complications, and hospital stay. The researchers conducted multivariable regression and sensitivity analyses to handle selection bias and confounding variables. Statistical analyses were performed with a significance level of p < 0.05. This study was approved by the Tekirdağ University Faculty of Medicine Ethics Committee (Approval No: 2024.22.02.06). Results: The C-group lesions showed proximity to the pleural surface at 5 mm compared to 8 mm (p = 0.048), indicating significant selection bias. Operation time was longer in the C-group (70 vs. 60 min, p = 0.115). Air leak duration and drainage time were similar between groups (p = 0.872, p = 0.176). Complication rates classified by Clavien–Dindo scale and hospital stay were comparable. The C-group showed reduced hospitalization expenses ($191.6 vs. $371.7) after adjusting for lesion characteristics and confounders while the clinical results between groups remained equivalent (adjusted OR for air leak: 0.68, 95% CI: 0.13–3.51, p = 0.645). The cost advantages persisted through sensitivity analysis which tested for selection bias effects. Conclusions: The clamp-and-suture method appears to offer a potentially cost-effective alternative to staplers for carefully selected peripheral lesions in VATS wedge resection, particularly in resource-limited settings. The preliminary results need to be treated as speculative because the study uses a non-randomized retrospective design with limited data from a small number of patients treated by one surgeon and shows evidence of selection bias. The obtained results do not qualify as practice-changing recommendations. The validation of these findings requires prospective randomized controlled trials with predetermined selection criteria and extended follow-up periods to establish clinical recommendations. Full article

Accurate segmentation of surgical instruments in endoscopic videos is crucial for robot-assisted surgery and intraoperative analysis. This paper presents a Segment-then-Classify framework that decouples mask generation from semantic classification to enhance spatial completeness and temporal stability. First, a Mask2Former-based segmentation backbone generates class-agnostic instance masks and region features. Then, a bounding box-guided instance-level spatiotemporal modeling module fuses geometric priors and temporal consistency through a lightweight transformer encoder. This design improves interpretability and robustness under occlusion and motion blur. Experiments on the EndoVis 2017 and 2018 datasets demonstrate that our framework achieves mIoU improvements of 3.06%, 2.99%, and 1.67% and mcIoU gains of 2.36%, 2.85%, and 6.06%, respectively, over previously state-of-the-art methods, while maintaining computational efficiency. Full article

Cataract surgery is the most common eye operation worldwide and is regarded as one of the safest procedures in medicine. Yet, despite its low complication rates, it generates a disproportionate share of litigation. The gap between excellent safety profiles and rising medico-legal claims is driven less by surgical outcomes than by patient expectations, often shaped by healthcare marketing and the promise of risk-free recovery. This narrative review explores the clinical and legal dimensions of cataract surgery, focusing on complications, perioperative risk factors, and medico-legal concepts of predictability and preventability. Particular emphasis is given to European frameworks, with the Italian Gelli-Bianco Law (Law No. 24/2017) providing a model of accountability that balances innovation and patient safety. Analysis shows that liability exposure spans all phases of surgery: preoperative (inadequate consent, poor documentation), intraoperative (posterior capsule rupture, zonular instability), and postoperative (endophthalmitis, poor follow-up). Practical strategies for risk reduction include advanced imaging such as macular OCT, rigorous adherence to updated guidelines, systematic video recording, and transparent perioperative communication. Patient-reported outcomes further highlight that satisfaction depends more on visual quality and dialogue than on spectacle independence. By translating legal principles into clinical strategies, this review offers surgeons actionable “surgical–legal pearls” to improve outcomes, strengthen patient trust, and reduce medico-legal vulnerability in high-volume cataract surgery. Full article

Objective: Non-intubated video-assisted thoracoscopic surgery (NIVATS) has emerged as a less invasive alternative to conventional intubated VATS (IVATS) for patients undergoing lobectomy for non-small cell lung cancer (NSCLC). However, concerns regarding its safety, efficacy, and oncologic adequacy remain. This meta-analysis aimed to compare perioperative and short-term outcomes between NIVATS and IVATS. Methods: A systematic review and meta-analysis were conducted in accordance with PRISMA guidelines. PubMed, Scopus, and Cochrane CENTRAL were searched through 30 June 2025. Studies comparing NIVATS and IVATS for anatomical lung resections (lobectomy and/or segmentectomy) in NSCLC were included; wedge resections were excluded. Primary endpoints included postoperative complications, operative time, intraoperative blood loss, lymph node yield, and 30-day mortality. Secondary endpoints were chest tube duration, hospital length of stay, anesthetic time, and conversion to thoracotomy rates. Risk of bias was assessed primarily with ROBINS-I; the Newcastle–Ottawa Scale was applied for sensitivity. Results: A total of seven studies (six retrospective and one randomized controlled trial) encompassing 851 patients (374 NIVATS, 477 IVATS) were included. NIVATS was associated with a significantly lower rate of postoperative complications (OR 0.50; 95% CI: 0.30–0.86; p = 0.01; I² = 0%), shorter operative time (minutes) (WMD −21.85; 95% CI: −38.49, −5.21; p = 0.01), anesthetic time (minutes) (WMD −4.62; 95% CI: −6.60, −2.65; p < 0.01), and reduced intraoperative blood loss (mL) (WMD −24.36; 95% CI: −30.67, −18.05; p < 0.01). There were no significant differences in lymph node yield or conversion to thoracotomy rates. No 30-day mortality was reported in either group. The quality of included studies was moderate, and publication bias was not evident. Conclusions: NIVATS appears to be a safe and effective alternative to IVATS in selected patients undergoing lobectomy for NSCLC. It offers improved perioperative outcomes without compromising surgical or oncologic standards. Prospective trials are needed to confirm these findings and assess long-term survival. Full article

Background/Objectives: Lidocaine has demonstrated immunomodulatory properties and promising antitumor effects in experimental models, but its impact on long-term outcomes following oncologic surgery remains unclear. This study aimed to compare the impact of intraoperative lidocaine versus remifentanil on long-term cancer outcomes after primary lung cancer surgery. Methods: This is a post hoc analysis of a randomized controlled trial (NCT03905837, EudraCT 2016-004271-52). From 154 patients who underwent elective lung resection via video-assisted thoracoscopic surgery (VATS) between January 2019 and June 2021 and were randomized to receive intraoperative lidocaine (intravenous or paravertebral) or remifentanil, we analyzed data from patients with confirmed primary lung cancer in the surgery specimen. Overall survival (OS) and disease-free survival (DFS) were assessed through May 2025. Survival outcomes were analyzed using Kaplan–Meier curves and log-rank tests. A multivariate Cox proportional hazards model was used to adjust for potential confounders. Results: Among the 97 patients with primary lung cancer finally included in the analysis, those in the lidocaine group exhibited improved OS compared with those who received intravenous remifentanil (log-rank p = 0.022). This association remained significant in the multivariate Cox regression analysis (HR 2.59, 95% CI 1.13–5.96, p = 0.025). No significant differences were observed in DFS overall (log-rank p = 0.283) or in DFS limited to recurrences of cancers present at the time of surgery, either the resected primary tumor or a prior malignancy (log-rank p = 0.080). Conclusions: In this post hoc analysis, lidocaine administration during oncologic lung resection was associated with improved OS in primary lung cancer patients. No differences in DFS were observed between groups; however, a non-significant trend toward improved DFS in lidocaine patients was noted when focusing on recurrences of cancers present at the time of surgery. Further investigation in larger prospective studies is warranted. Full article

Background/Objectives: Colorectal anastomotic leak (CAL) is one of the most severe postoperative complications in colorectal surgery, impacting patient morbidity and mortality. Current risk assessment methods rely on clinical and intraoperative factors, but no real-time predictive tool exists. This study aimed to develop an artificial intelligence model based on intraoperative laparoscopic recording of the anastomosis for CAL prediction. Methods: A convolutional neural network (CNN) was trained with annotated frames from colorectal surgery videos across three international high-volume centers (Instituto Português de Oncologia de Lisboa, Hospital das Clínicas de Ribeirão Preto, and Royal Liverpool University Hospital). The dataset included a total of 5356 frames from 26 patients, 2007 with CAL and 3349 showing normal anastomosis. Four CNN architectures (EfficientNetB0, EfficientNetB7, ResNet50, and MobileNetV2) were tested. The models’ performance was evaluated using their sensitivity, specificity, accuracy, and area under the receiver operating characteristic (AUROC) curve. Heatmaps were generated to identify key image regions influencing predictions. Results: The best-performing model achieved an accuracy of 99.6%, AUROC of 99.6%, sensitivity of 99.2%, specificity of 100.0%, PPV of 100.0%, and NPV of 98.9%. The model reliably identified CAL-positive frames and provided visual explanations through heatmaps. Conclusions: To our knowledge, this is the first AI model developed to predict CAL using intraoperative video analysis. Its accuracy suggests the potential to redefine surgical decision-making by providing real-time risk assessment. Further refinement with a larger dataset and diverse surgical techniques could enable intraoperative interventions to prevent CAL before it occurs, marking a paradigm shift in colorectal surgery. Full article

Background/Objectives: The extensive use of computed tomography (CT) has led to a significant increase in the detection of small and non-palpable pulmonary nodules, necessitating the use of invasive methods for definitive diagnosis. Video-assisted thoracoscopic surgery (VATS) has become the preferred procedure for nodule resections; however, intraoperative localization remains challenging, especially for deep or subsolid lesions. This study explores whether SPECT/CT improves the technical and clinical outcomes of radio-guided occult lesion localization (ROLL) before uniportal video-assisted thoracoscopic surgery (u-VATS). Methods: This is a retrospective study involving consecutive patients referred for the resection of pulmonary nodules who underwent CT-guided ROLL followed by u-VATS between September 2017 and December 2024. From January 2023, SPECT/CT was systematically added after planar imaging. The cohort was divided into a planar group and a planar + SPECT/CT group. The inclusion criteria involved nodules sized ≤ 2 cm, with ground glass or solid characteristics, located at a depth of <6 cm from the pleural surface. ^99mTc-MAA injected activity, timing, the classification of planar and SPECT/CT image findings (focal uptake, multisite with focal uptake, multisite without focal uptake), spillage, and post-procedure complications were evaluated. Statistical analysis was performed, with continuous data expressed as the median and categorical data as the number. Comparisons were made using chi-square tests for categorical variables and the Mann–Whitney U test for procedural duration. Cohen’s kappa coefficient was calculated to assess agreement between imaging modalities. Results: In total, 125 patients were selected for CT-guided radiotracer injection followed by uniportal-VATS. The planar group and planar + SPECT/CT group comprised 60 and 65 patients, respectively. Focal uptake was detected in 68 (54%), multisite with focal uptake in 46 (36.8%), and multisite without focal uptake in 11 patients (8.8%). In comparative analyses between planar and SPECT/CT imaging in 65 patients, 91% exhibited focal uptake, revealing significant differences in classification for 40% of the patients. SPECT/CT corrected the classification of 23 patients initially categorized as multisite with focal uptake to focal uptake, improving localization accuracy. The mean procedure duration was 39 min with SPECT/CT. Pneumothorax was more frequently detected with SPECT/CT (43% vs. 1.6%). The intraoperative localization success rate was 96%. Conclusions: SPECT/CT imaging in the ROLL procedure for detecting pulmonary nodules before u-VATs demonstrates a significant advantage in reclassifying radiotracer positioning compared to planar imaging. Considering its limited impact on surgical success rates and additional procedural time, SPECT/CT should be reserved for technically challenging cases. Larger sample sizes, multicentric and prospective randomized studies, and formal cost–utility analyses are warranted. Full article

Background: Previous studies have shown that ventilation strategies used in general anesthesia influence perioperative outcomes of video-assisted thoracoscopic surgery (VATS). This study investigated the perioperative effects of non-intubated anesthesia (NIA) versus intubated anesthesia (IA) in patients with early-stage non-small cell lung cancer (NSCLC) undergoing VATS. Methods: This retrospective cohort study analyzed patients who underwent elective VATS for early-stage NSCLC between January 2015 and December 2022. Patients were categorized into the NIA and IA groups based on the ventilation strategies during general anesthesia. Comprehensive outcome data, including intraoperative and postoperative variables, were compared between the two groups. Univariate and multivariate logistic regression models were used to assess the odds ratios for conversion from NIA to IA. Results: A total of 372 patients who received NIA and 1560 who received IA for VATS were eligible for analysis. After propensity score matching, 336 patients were included in each group. In the matched analysis, patients who received NIA demonstrated favorable perioperative outcomes, including reduced opioid consumption, lower postoperative complication rates, and shorter hospital stays, compared to those who received IA. Additionally, patients with a lower baseline oxygen saturation and those who experienced intraoperative pulmonary and cardiovascular adverse events had a higher risk of conversion from NIA to IA. Conclusions: NIA during VATS in patients with early-stage NSCLC was associated with superior perioperative outcomes. Prospective studies are warranted to further evaluate the impact of NIA on perioperative outcomes in this patient population. Full article

Background/Objectives: Primary hyperparathyroidism (PHPT) is a prevalent endocrine disorder, with ectopic mediastinal parathyroid adenomas accounting for up to 30% of cases, posing significant diagnostic and surgical challenges. While traditional management relies on invasive procedures, minimally invasive techniques such as video-assisted thoracoscopic surgery (VATS) have emerged as viable alternatives. This study addresses a gap in the current literature by presenting our experience with VATS for mediastinal ectopic parathyroid adenomas, particularly in underreported retrotracheal/paraesophageal locations. By integrating a retrospective case series with a systematic literature review, we highlight evolving surgical strategies and their implications for patient outcomes in anatomically complex cases. Methods: A retrospective analysis was conducted over a three-year period on patients diagnosed with mediastinal ectopic parathyroid adenomas. Data on demographic characteristics, preoperative imaging, surgical techniques, intraoperative findings, and postoperative outcomes were collected. This study primarily compared the outcomes of VATS with those of traditional thoracotomy, with a focus on surgical success, complication rates, and length of hospital stay. Results: Six patients underwent surgical resection for mediastinal ectopic parathyroid adenomas (two intrahymic and four retrotracheal/paraesophgeal). VATS was the preferred approach in all cases, with one patient requiring conversion to thoracotomy due to challenging vascular anatomy. Surgical success, defined as the normalization of postoperative serum calcium levels, was achieved in all cases. The median operative time was 80 min, and the mean hospital stay was 6.25 days. One patient developed transient postoperative hypocalcemia, necessitating supplementation. No major surgical complications were observed. Conclusions: This study supports VATS as a safe and effective approach for mediastinal ectopic parathyroid adenoma resection, offering reduced morbidity and shorter recovery times compared to traditional open surgery. The findings align with emerging evidence advocating for minimally invasive techniques in complex mediastinal surgeries. Full article

Purpose: Augmented reality (AR) may allow vitreoretinal surgeons to leverage microscope-integrated digital imaging systems to analyze and highlight key retinal anatomic features in real time, possibly improving safety and precision during surgery. By employing convolutional neural networks (CNNs) for retina vessel segmentation, a retinal coordinate system can be created that allows pre-operative images of capillary non-perfusion or retinal breaks to be digitally aligned and overlayed upon the surgical field in real time. Such technology may be useful in assuring thorough laser treatment of capillary non-perfusion or in using pre-operative optical coherence tomography (OCT) to guide macular surgery when microscope-integrated OCT (MIOCT) is not available. Methods: This study is a retrospective analysis involving the development and testing of a novel image-registration algorithm for vitreoretinal surgery. Fifteen anonymized cases of pars plana vitrectomy with epiretinal membrane peeling, along with corresponding preoperative fundus photographs and optical coherence tomography (OCT) images, were retrospectively collected from the Mayo Clinic database. We developed a TPU (Tensor-Processing Unit)-accelerated CNN for semantic segmentation of retinal vessels from fundus photographs and subsequent real-time image registration in surgical video streams. An iterative patch-wise cross-correlation (IPCC) algorithm was developed for image registration, with a focus on optimizing processing speeds and maintaining high spatial accuracy. The primary outcomes measured were processing speed in frames per second (FPS) and the spatial accuracy of image registration, quantified by the Dice coefficient between registered and manually aligned images. Results: When deployed on an Edge TPU, the CNN model combined with our image-registration algorithm processed video streams at a rate of 14 FPS, which is superior to processing rates achieved on other standard hardware configurations. The IPCC algorithm efficiently aligned pre-operative and intraoperative images, showing high accuracy in comparison to manual registration. Conclusions: This study demonstrates the feasibility of using TPU-accelerated CNNs for enhanced AR in vitreoretinal surgery. Full article