SeeSaw: Learning Soft Tissue Deformation from Laparoscopy Videos with GNNs.

A major challenge in image-guided laparoscopic surgery is that structures of interest often deform and go, even if only momentarily, out of view. Methods which rely on having an up-to-date impression of those structures, such as registration or localisation, are undermined in these circumstances. This is particularly true for soft-tissue structures that continually change shape - in registration, they must often be re-mapped. Furthermore, methods which require 'revisiting' of previously seen areas cannot in principle function reliably in dynamic contexts, drastically weakening their uptake in the operating room. We present a novel approach for learning to estimate the deformed states of previously seen soft tissue surfaces from currently observable regions, using a combined approach that includes a Graph Neural Network (GNN). The training data is based on stereo laparoscopic surgery videos, generated semi-automatically with minimal labelling effort. Trackable segments are first identified using a feature detection algorithm, from which surface meshes are produced using depth estimation and delaunay triangulation. We show the method can predict the displacements of previously visible soft tissue structures connected to currently visible regions with observed displacements, both on our own data and porcine data. Our innovative approach learns to compensate non-rigidity in abdominal endoscopic scenes directly from stereo laparoscopic videos through targeting a new problem formulation, and stands to benefit a variety of target applications in dynamic environments. Project page for this work: https://gitlab.com/nct_tso_public/seesaw-soft-tissue-deformation.

Can you feel the force just right? Tactile force feedback for training of minimally invasive surgery-evaluation of vibration feedback for adequate force application.

BACKGROUND: Tissue handling is a crucial skill for surgeons and is challenging to learn. The aim of this study was to develop laparoscopic instruments with different integrated tactile vibration feedback by varying different tactile modalities and assess its effect on tissue handling skills. METHODS: Standard laparoscopic instruments were equipped with a vibration effector, which was controlled by a microcomputer attached to a force sensor platform. One of three different vibration feedbacks (F1: double vibration > 2 N; F2: increasing vibration relative to force; F3: one vibration > 1.5 N and double vibration > 2 N) was applied to the instruments. In this multicenter crossover trial, surgical novices and expert surgeons performed two laparoscopic tasks (Peg transfer, laparoscopic suture, and knot) each with all the three vibration feedback modalities and once without any feedback, in a randomized order. The primary endpoint was force exertion. RESULTS: A total of 57 subjects (15 surgeons, 42 surgical novices) were included in the trial. In the Peg transfer task, there were no differences between the tactile feedback modalities in terms of force application. However, in subgroup analysis, the use of F2 resulted in a significantly lower mean-force application (p-value = 0.02) among the student group. In the laparoscopic suture and knot task, all participants exerted significantly lower mean and peak forces using F2 (p-value < 0.01). These findings remained significant after subgroup analysis for both, the student and surgeon groups individually. The condition without tactile feedback led to the highest mean and peak force exertion compared to the three other feedback modalities. CONCLUSION: Continuous tactile vibration feedback decreases the mean and peak force applied during laparoscopic training tasks. This effect is more pronounced in demanding tasks such as laparoscopic suturing and knot tying and might be more beneficial for students. Laparoscopic tasks without feedback lead to increased force application.

The development of tissue handling skills is sufficient and comparable after training in virtual reality or on a surgical robotic system: a prospective randomized trial.

BACKGROUND: Virtual reality is a frequently chosen method for learning the basics of robotic surgery. However, it is unclear whether tissue handling is adequately trained in VR training compared to training on a real robotic system. METHODS: In this randomized controlled trial, participants were split into two groups for "Fundamentals of Robotic Surgery (FRS)" training on either a DaVinci VR simulator (VR group) or a DaVinci robotic system (Robot group). All participants completed four tasks on the DaVinci robotic system before training (Baseline test), after proficiency in three FRS tasks (Midterm test), and after proficiency in all FRS tasks (Final test). Primary endpoints were forces applied across tests. RESULTS: This trial included 87 robotic novices, of which 43 and 44 participants received FRS training in VR group and Robot group, respectively. The Baseline test showed no significant differences in force application between the groups indicating a sufficient randomization. In the Midterm and Final test, the force application was not different between groups. Both groups displayed sufficient learning curves with significant improvement of force application. However, the Robot group needed significantly less repetitions in the three FRS tasks Ring tower (Robot: 2.48 vs. VR: 5.45; p < 0.001), Knot Tying (Robot: 5.34 vs. VR: 8.13; p = 0.006), and Vessel Energy Dissection (Robot: 2 vs. VR: 2.38; p = 0.001) until reaching proficiency. CONCLUSION: Robotic tissue handling skills improve significantly and comparably after both VR training and training on a real robotic system, but training on a VR simulator might be less efficient.

One model to use them all: training a segmentation model with complementary datasets.

PURPOSE: Understanding surgical scenes is crucial for computer-assisted surgery systems to provide intelligent assistance functionality. One way of achieving this is via scene segmentation using machine learning (ML). However, such ML models require large amounts of annotated training data, containing examples of all relevant object classes, which are rarely available. In this work, we propose a method to combine multiple partially annotated datasets, providing complementary annotations, into one model, enabling better scene segmentation and the use of multiple readily available datasets. METHODS: Our method aims to combine available data with complementary labels by leveraging mutual exclusive properties to maximize information. Specifically, we propose to use positive annotations of other classes as negative samples and to exclude background pixels of these binary annotations, as we cannot tell if a positive prediction by the model is correct. RESULTS: We evaluate our method by training a DeepLabV3 model on the publicly available Dresden Surgical Anatomy Dataset, which provides multiple subsets of binary segmented anatomical structures. Our approach successfully combines 6 classes into one model, significantly increasing the overall Dice Score by 4.4% compared to an ensemble of models trained on the classes individually. By including information on multiple classes, we were able to reduce the confusion between classes, e.g. a 24% drop for stomach and colon. CONCLUSION: By leveraging multiple datasets and applying mutual exclusion constraints, we developed a method that improves surgical scene segmentation performance without the need for fully annotated datasets. Our results demonstrate the feasibility of training a model on multiple complementary datasets. This paves the way for future work further alleviating the need for one specialized large, fully segmented dataset but instead the use of already existing datasets.

On the pitfalls of Batch Normalization for end-to-end video learning: A study on surgical workflow analysis.

Batch Normalization's (BN) unique property of depending on other samples in a batch is known to cause problems in several tasks, including sequence modeling. Yet, BN-related issues are hardly studied for long video understanding, despite the ubiquitous use of BN in CNNs (Convolutional Neural Networks) for feature extraction. Especially in surgical workflow analysis, where the lack of pretrained feature extractors has led to complex, multi-stage training pipelines, limited awareness of BN issues may have hidden the benefits of training CNNs and temporal models end to end. In this paper, we analyze pitfalls of BN in video learning, including issues specific to online tasks such as a 'cheating' effect in anticipation. We observe that BN's properties create major obstacles for end-to-end learning. However, using BN-free backbones, even simple CNN-LSTMs beat the state of the art on three surgical workflow benchmarks by utilizing adequate end-to-end training strategies which maximize temporal context. We conclude that awareness of BN's pitfalls is crucial for effective end-to-end learning in surgical tasks. By reproducing results on natural-video datasets, we hope our insights will benefit other areas of video learning as well. Code is available at: https://gitlab.com/nct_tso_public/pitfalls_bn.

[The digital operating room : Chances and risks of artificial intelligence]. / Der digitale Operationssaal : Chancen und Risiken künstlicher Intelligenz.

At the central workplace of the surgeon the digitalization of the operating room has particular consequences for the surgical work. Starting with intraoperative cross-sectional imaging and sonography, through functional imaging, minimally invasive and robot-assisted surgery up to digital surgical and anesthesiological documentation, the vast majority of operating rooms are now at least partially digitalized. The increasing digitalization of the whole process chain enables not only for the collection but also the analysis of big data. Current research focuses on artificial intelligence for the analysis of intraoperative data as the prerequisite for assistance systems that support surgical decision making or warn of risks; however, these technologies raise new ethical questions for the surgical community that affect the core of surgical work.

Comparing a virtual reality head-mounted display to on-screen three-dimensional visualization and two-dimensional computed tomography data for training in decision making in hepatic surgery: a randomized controlled study.

OBJECTIVE: Evaluation of the benefits of a virtual reality (VR) environment with a head-mounted display (HMD) for decision-making in liver surgery. BACKGROUND: Training in liver surgery involves appraising radiologic images and considering the patient's clinical information. Accurate assessment of 2D-tomography images is complex and requires considerable experience, and often the images are divorced from the clinical information. We present a comprehensive and interactive tool for visualizing operation planning data in a VR environment using a head-mounted-display and compare it to 3D visualization and 2D-tomography. METHODS: Ninety medical students were randomized into three groups (1:1:1 ratio). All participants analyzed three liver surgery patient cases with increasing difficulty. The cases were analyzed using 2D-tomography data (group "2D"), a 3D visualization on a 2D display (group "3D") or within a VR environment (group "VR"). The VR environment was displayed using the "Oculus Rift ™" HMD technology. Participants answered 11 questions on anatomy, tumor involvement and surgical decision-making and 18 evaluative questions (Likert scale). RESULTS: Sum of correct answers were significantly higher in the 3D (7.1 ± 1.4, p < 0.001) and VR (7.1 ± 1.4, p < 0.001) groups than the 2D group (5.4 ± 1.4) while there was no difference between 3D and VR (p = 0.987). Times to answer in the 3D (6:44 ± 02:22 min, p < 0.001) and VR (6:24 ± 02:43 min, p < 0.001) groups were significantly faster than the 2D group (09:13 ± 03:10 min) while there was no difference between 3D and VR (p = 0.419). The VR environment was evaluated as most useful for identification of anatomic anomalies, risk and target structures and for the transfer of anatomical and pathological information to the intraoperative situation in the questionnaire. CONCLUSIONS: A VR environment with 3D visualization using a HMD is useful as a surgical training tool to accurately and quickly determine liver anatomy and tumor involvement in surgery.

AIxSuture: vision-based assessment of open suturing skills.

PURPOSE: Efficient and precise surgical skills are essential in ensuring positive patient outcomes. By continuously providing real-time, data driven, and objective evaluation of surgical performance, automated skill assessment has the potential to greatly improve surgical skill training. Whereas machine learning-based surgical skill assessment is gaining traction for minimally invasive techniques, this cannot be said for open surgery skills. Open surgery generally has more degrees of freedom when compared to minimally invasive surgery, making it more difficult to interpret. In this paper, we present novel approaches for skill assessment for open surgery skills. METHODS: We analyzed a novel video dataset for open suturing training. We provide a detailed analysis of the dataset and define evaluation guidelines, using state of the art deep learning models. Furthermore, we present novel benchmarking results for surgical skill assessment in open suturing. The models are trained to classify a video into three skill levels based on the global rating score. To obtain initial results for video-based surgical skill classification, we benchmarked a temporal segment network with both an I3D and a Video Swin backbone on this dataset. RESULTS: The dataset is composed of 314 videos of approximately five minutes each. Model benchmarking results are an accuracy and F1 score of up to 75 and 72%, respectively. This is similar to the performance achieved by the individual raters, regarding inter-rater agreement and rater variability. We present the first end-to-end trained approach for skill assessment for open surgery training. CONCLUSION: We provide a thorough analysis of a new dataset as well as novel benchmarking results for surgical skill assessment. This opens the doors to new advances in skill assessment by enabling video-based skill assessment for classic surgical techniques with the potential to improve the surgical outcome of patients.

The Dresden in vivo OCT dataset for automatic middle ear segmentation.

Endoscopic optical coherence tomography (OCT) offers a non-invasive approach to perform the morphological and functional assessment of the middle ear in vivo. However, interpreting such OCT images is challenging and time-consuming due to the shadowing of preceding structures. Deep neural networks have emerged as a promising tool to enhance this process in multiple aspects, including segmentation, classification, and registration. Nevertheless, the scarcity of annotated datasets of OCT middle ear images poses a significant hurdle to the performance of neural networks. We introduce the Dresden in vivo OCT Dataset of the Middle Ear (DIOME) featuring 43 OCT volumes from both healthy and pathological middle ears of 29 subjects. DIOME provides semantic segmentations of five crucial anatomical structures (tympanic membrane, malleus, incus, stapes and promontory), and sparse landmarks delineating the salient features of the structures. The availability of these data facilitates the training and evaluation of algorithms regarding various analysis tasks with middle ear OCT images, e.g. diagnostics.

Exploring semantic consistency in unpaired image translation to generate data for surgical applications.

PURPOSE: In surgical computer vision applications, data privacy and expert annotation challenges impede the acquisition of labeled training data. Unpaired image-to-image translation techniques have been explored to automatically generate annotated datasets by translating synthetic images into a realistic domain. The preservation of structure and semantic consistency, i.e., per-class distribution during translation, poses a significant challenge, particularly in cases of semantic distributional mismatch. METHOD: This study empirically investigates various translation methods for generating data in surgical applications, explicitly focusing on semantic consistency. Through our analysis, we introduce a novel and simple combination of effective approaches, which we call ConStructS. The defined losses within this approach operate on multiple image patches and spatial resolutions during translation. RESULTS: Various state-of-the-art models were extensively evaluated on two challenging surgical datasets. With two different evaluation schemes, the semantic consistency and the usefulness of the translated images on downstream semantic segmentation tasks were evaluated. The results demonstrate the effectiveness of the ConStructS method in minimizing semantic distortion, with images generated by this model showing superior utility for downstream training. CONCLUSION: In this study, we tackle semantic inconsistency in unpaired image translation for surgical applications with minimal labeled data. The simple model (ConStructS) enhances consistency during translation and serves as a practical way of generating fully labeled and semantically consistent datasets at minimal cost. Our code is available at https://gitlab.com/nct_tso_public/constructs .

Non-rigid point cloud registration for middle ear diagnostics with endoscopic optical coherence tomography.

PURPOSE: Middle ear infection is the most prevalent inflammatory disease, especially among the pediatric population. Current diagnostic methods are subjective and depend on visual cues from an otoscope, which is limited for otologists to identify pathology. To address this shortcoming, endoscopic optical coherence tomography (OCT) provides both morphological and functional in vivo measurements of the middle ear. However, due to the shadow of prior structures, interpretation of OCT images is challenging and time-consuming. To facilitate fast diagnosis and measurement, improvement in the readability of OCT data is achieved by merging morphological knowledge from ex vivo middle ear models with OCT volumetric data, so that OCT applications can be further promoted in daily clinical settings. METHODS: We propose C2P-Net: a two-staged non-rigid registration pipeline for complete to partial point clouds, which are sampled from ex vivo and in vivo OCT models, respectively. To overcome the lack of labeled training data, a fast and effective generation pipeline in Blender3D is designed to simulate middle ear shapes and extract in vivo noisy and partial point clouds. RESULTS: We evaluate the performance of C2P-Net through experiments on both synthetic and real OCT datasets. The results demonstrate that C2P-Net is generalized to unseen middle ear point clouds and capable of handling realistic noise and incompleteness in synthetic and real OCT data. CONCLUSIONS: In this work, we aim to enable diagnosis of middle ear structures with the assistance of OCT images. We propose C2P-Net: a two-staged non-rigid registration pipeline for point clouds to support the interpretation of in vivo noisy and partial OCT images for the first time. Code is available at: https://gitlab.com/nct_tso_public/c2p-net.

Coming in handy: CeTI-Age - A comprehensive database of kinematic hand movements across the lifespan.

The Tactile Internet aims to advance human-human and human-machine interactions that also utilize hand movements in real, digitized, and remote environments. Attention to elderly generations is necessary to make the Tactile Internet age inclusive. We present the first age-representative kinematic database consisting of various hand gesturing and grasping movements at individualized paces, thus capturing naturalistic movements. We make this comprehensive database of kinematic hand movements across the adult lifespan (CeTI-Age-Kinematic-Hand) publicly available to facilitate a deeper understanding of intra-individual-focusing especially on age-related differences-and inter-individual variability in hand kinematics. The core of the database contains participants' hand kinematics recorded with wearable resistive bend sensors, individual static 3D hand models, and all instructional videos used during the data acquisition. Sixty-three participants ranging from age 20 to 80 years performed six repetitions of 40 different naturalistic hand movements at individual paces. This unique database with data recorded from an adult lifespan sample can be used to advance machine-learning approaches in hand kinematic modeling and movement prediction for age-inclusive applications.

Active learning for extracting surgomic features in robot-assisted minimally invasive esophagectomy: a prospective annotation study.

BACKGROUND: With Surgomics, we aim for personalized prediction of the patient's surgical outcome using machine-learning (ML) on multimodal intraoperative data to extract surgomic features as surgical process characteristics. As high-quality annotations by medical experts are crucial, but still a bottleneck, we prospectively investigate active learning (AL) to reduce annotation effort and present automatic recognition of surgomic features. METHODS: To establish a process for development of surgomic features, ten video-based features related to bleeding, as highly relevant intraoperative complication, were chosen. They comprise the amount of blood and smoke in the surgical field, six instruments, and two anatomic structures. Annotation of selected frames from robot-assisted minimally invasive esophagectomies was performed by at least three independent medical experts. To test whether AL reduces annotation effort, we performed a prospective annotation study comparing AL with equidistant sampling (EQS) for frame selection. Multiple Bayesian ResNet18 architectures were trained on a multicentric dataset, consisting of 22 videos from two centers. RESULTS: In total, 14,004 frames were tag annotated. A mean F1-score of 0.75 ± 0.16 was achieved for all features. The highest F1-score was achieved for the instruments (mean 0.80 ± 0.17). This result is also reflected in the inter-rater-agreement (1-rater-kappa > 0.82). Compared to EQS, AL showed better recognition results for the instruments with a significant difference in the McNemar test comparing correctness of predictions. Moreover, in contrast to EQS, AL selected more frames of the four less common instruments (1512 vs. 607 frames) and achieved higher F1-scores for common instruments while requiring less training frames. CONCLUSION: We presented ten surgomic features relevant for bleeding events in esophageal surgery automatically extracted from surgical video using ML. AL showed the potential to reduce annotation effort while keeping ML performance high for selected features. The source code and the trained models are published open source.

Anatomy segmentation in laparoscopic surgery: comparison of machine learning and human expertise - an experimental study.

BACKGROUND: Lack of anatomy recognition represents a clinically relevant risk in abdominal surgery. Machine learning (ML) methods can help identify visible patterns and risk structures; however, their practical value remains largely unclear. MATERIALS AND METHODS: Based on a novel dataset of 13 195 laparoscopic images with pixel-wise segmentations of 11 anatomical structures, we developed specialized segmentation models for each structure and combined models for all anatomical structures using two state-of-the-art model architectures (DeepLabv3 and SegFormer) and compared segmentation performance of algorithms to a cohort of 28 physicians, medical students, and medical laypersons using the example of pancreas segmentation. RESULTS: Mean Intersection-over-Union for semantic segmentation of intra-abdominal structures ranged from 0.28 to 0.83 and from 0.23 to 0.77 for the DeepLabv3-based structure-specific and combined models, and from 0.31 to 0.85 and from 0.26 to 0.67 for the SegFormer-based structure-specific and combined models, respectively. Both the structure-specific and the combined DeepLabv3-based models are capable of near-real-time operation, while the SegFormer-based models are not. All four models outperformed at least 26 out of 28 human participants in pancreas segmentation. CONCLUSIONS: These results demonstrate that ML methods have the potential to provide relevant assistance in anatomy recognition in minimally invasive surgery in near-real-time. Future research should investigate the educational value and subsequent clinical impact of the respective assistance systems.

Artificial Intelligence for context-aware surgical guidance in complex robot-assisted oncological procedures: An exploratory feasibility study.

INTRODUCTION: Complex oncological procedures pose various surgical challenges including dissection in distinct tissue planes and preservation of vulnerable anatomical structures throughout different surgical phases. In rectal surgery, violation of dissection planes increases the risk of local recurrence and autonomous nerve damage resulting in incontinence and sexual dysfunction. This work explores the feasibility of phase recognition and target structure segmentation in robot-assisted rectal resection (RARR) using machine learning. MATERIALS AND METHODS: A total of 57 RARR were recorded and subsets of these were annotated with respect to surgical phases and exact locations of target structures (anatomical structures, tissue types, static structures, and dissection areas). For surgical phase recognition, three machine learning models were trained: LSTM, MSTCN, and Trans-SVNet. Based on pixel-wise annotations of target structures in 9037 images, individual segmentation models based on DeepLabv3 were trained. Model performance was evaluated using F1 score, Intersection-over-Union (IoU), accuracy, precision, recall, and specificity. RESULTS: The best results for phase recognition were achieved with the MSTCN model (F1 score: 0.82 ± 0.01, accuracy: 0.84 ± 0.03). Mean IoUs for target structure segmentation ranged from 0.14 ± 0.22 to 0.80 ± 0.14 for organs and tissue types and from 0.11 ± 0.11 to 0.44 ± 0.30 for dissection areas. Image quality, distorting factors (i.e. blood, smoke), and technical challenges (i.e. lack of depth perception) considerably impacted segmentation performance. CONCLUSION: Machine learning-based phase recognition and segmentation of selected target structures are feasible in RARR. In the future, such functionalities could be integrated into a context-aware surgical guidance system for rectal surgery.

SAGES consensus recommendations on surgical video data use, structure, and exploration (for research in artificial intelligence, clinical quality improvement, and surgical education).

BACKGROUND: Surgery generates a vast amount of data from each procedure. Particularly video data provides significant value for surgical research, clinical outcome assessment, quality control, and education. The data lifecycle is influenced by various factors, including data structure, acquisition, storage, and sharing; data use and exploration, and finally data governance, which encompasses all ethical and legal regulations associated with the data. There is a universal need among stakeholders in surgical data science to establish standardized frameworks that address all aspects of this lifecycle to ensure data quality and purpose. METHODS: Working groups were formed, among 48 representatives from academia and industry, including clinicians, computer scientists and industry representatives. These working groups focused on: Data Use, Data Structure, Data Exploration, and Data Governance. After working group and panel discussions, a modified Delphi process was conducted. RESULTS: The resulting Delphi consensus provides conceptualized and structured recommendations for each domain related to surgical video data. We identified the key stakeholders within the data lifecycle and formulated comprehensive, easily understandable, and widely applicable guidelines for data utilization. Standardization of data structure should encompass format and quality, data sources, documentation, metadata, and account for biases within the data. To foster scientific data exploration, datasets should reflect diversity and remain adaptable to future applications. Data governance must be transparent to all stakeholders, addressing legal and ethical considerations surrounding the data. CONCLUSION: This consensus presents essential recommendations around the generation of standardized and diverse surgical video databanks, accounting for multiple stakeholders involved in data generation and use throughout its lifecycle. Following the SAGES annotation framework, we lay the foundation for standardization of data use, structure, and exploration. A detailed exploration of requirements for adequate data governance will follow.

Does practice make perfect? Laparoscopic training mainly improves motion efficiency: a prospective trial.

Training improves skills in minimally invasive surgery. This study aimed to investigate the learning curves of complex motion parameters for both hands during a standardized training course using a novel measurement tool. An additional focus was placed on the parameters representing surgical safety and precision. Fifty-six laparoscopic novices participated in a training course on the basic skills of minimally invasive surgery based on a modified Fundamentals of Laparoscopic Surgery (FLS) curriculum. Before, twice during, and once after the practical lessons, all participants had to perform four laparoscopic tasks (peg transfer, precision cut, balloon resection, and laparoscopic suture and knot), which were recorded and analyzed using an instrument motion analysis system. Participants significantly improved the time per task for all four tasks (all p < 0.001). The individual instrument path length decreased significantly for the dominant and non-dominant hands in all four tasks. Similarly, both hands became significantly faster in all tasks, with the exception of the non-dominant hand in the precision cut task. In terms of relative idle time, only in the peg transfer task did both hands improve significantly, while in the precision cut task, only the dominant hand performed better. In contrast, the motion volume of both hands combined was reduced in only one task (precision cut, p = 0.01), whereas no significant improvement in the relative time of instruments being out of view was observed. FLS-based skills training increases motion efficiency primarily by increasing speed and reducing idle time and path length. Parameters relevant for surgical safety and precision (motion volume and relative time of instruments being out of view) are minimally affected by short-term training. Consequently, surgical training should also focus on safety and precision-related parameters, and assessment of these parameters should be incorporated into basic skill training accordingly.

Prediction of clinically relevant postoperative pancreatic fistula using radiomic features and preoperative data.

Clinically relevant postoperative pancreatic fistula (CR-POPF) can significantly affect the treatment course and outcome in pancreatic cancer patients. Preoperative prediction of CR-POPF can aid the surgical decision-making process and lead to better perioperative management of patients. In this retrospective study of 108 pancreatic head resection patients, we present risk models for the prediction of CR-POPF that use combinations of preoperative computed tomography (CT)-based radiomic features, mesh-based volumes of annotated intra- and peripancreatic structures and preoperative clinical data. The risk signatures were evaluated and analysed in detail by visualising feature expression maps and by comparing significant features to the established CR-POPF risk measures. Out of the risk models that were developed in this study, the combined radiomic and clinical signature performed best with an average area under receiver operating characteristic curve (AUC) of 0.86 and a balanced accuracy score of 0.76 on validation data. The following pre-operative features showed significant correlation with outcome in this signature ([Formula: see text]) - texture and morphology of the healthy pancreatic segment, intensity volume histogram-based feature of the pancreatic duct segment, morphology of the combined segment, and BMI. The predictions of this pre-operative signature showed strong correlation (Spearman correlation co-efficient, [Formula: see text]) with the intraoperative updated alternative fistula risk score (ua-FRS), which is the clinical gold standard for intraoperative CR-POPF risk stratification. These results indicate that the proposed combined radiomic and clinical signature developed solely based on preoperatively available clinical and routine imaging data can perform on par with the current state-of-the-art intraoperative models for CR-POPF risk stratification.

Comparative validation of machine learning algorithms for surgical workflow and skill analysis with the HeiChole benchmark.

PURPOSE: Surgical workflow and skill analysis are key technologies for the next generation of cognitive surgical assistance systems. These systems could increase the safety of the operation through context-sensitive warnings and semi-autonomous robotic assistance or improve training of surgeons via data-driven feedback. In surgical workflow analysis up to 91% average precision has been reported for phase recognition on an open data single-center video dataset. In this work we investigated the generalizability of phase recognition algorithms in a multicenter setting including more difficult recognition tasks such as surgical action and surgical skill. METHODS: To achieve this goal, a dataset with 33 laparoscopic cholecystectomy videos from three surgical centers with a total operation time of 22 h was created. Labels included framewise annotation of seven surgical phases with 250 phase transitions, 5514 occurences of four surgical actions, 6980 occurences of 21 surgical instruments from seven instrument categories and 495 skill classifications in five skill dimensions. The dataset was used in the 2019 international Endoscopic Vision challenge, sub-challenge for surgical workflow and skill analysis. Here, 12 research teams trained and submitted their machine learning algorithms for recognition of phase, action, instrument and/or skill assessment. RESULTS: F1-scores were achieved for phase recognition between 23.9% and 67.7% (n = 9 teams), for instrument presence detection between 38.5% and 63.8% (n = 8 teams), but for action recognition only between 21.8% and 23.3% (n = 5 teams). The average absolute error for skill assessment was 0.78 (n = 1 team). CONCLUSION: Surgical workflow and skill analysis are promising technologies to support the surgical team, but there is still room for improvement, as shown by our comparison of machine learning algorithms. This novel HeiChole benchmark can be used for comparable evaluation and validation of future work. In future studies, it is of utmost importance to create more open, high-quality datasets in order to allow the development of artificial intelligence and cognitive robotics in surgery.

The Dresden Surgical Anatomy Dataset for Abdominal Organ Segmentation in Surgical Data Science.

Laparoscopy is an imaging technique that enables minimally-invasive procedures in various medical disciplines including abdominal surgery, gynaecology and urology. To date, publicly available laparoscopic image datasets are mostly limited to general classifications of data, semantic segmentations of surgical instruments and low-volume weak annotations of specific abdominal organs. The Dresden Surgical Anatomy Dataset provides semantic segmentations of eight abdominal organs (colon, liver, pancreas, small intestine, spleen, stomach, ureter, vesicular glands), the abdominal wall and two vessel structures (inferior mesenteric artery, intestinal veins) in laparoscopic view. In total, this dataset comprises 13195 laparoscopic images. For each anatomical structure, we provide over a thousand images with pixel-wise segmentations. Annotations comprise semantic segmentations of single organs and one multi-organ-segmentation dataset including segments for all eleven anatomical structures. Moreover, we provide weak annotations of organ presence for every single image. This dataset markedly expands the horizon for surgical data science applications of computer vision in laparoscopic surgery and could thereby contribute to a reduction of risks and faster translation of Artificial Intelligence into surgical practice.