Instructional cataract surgery videos are crucial for ophthalmologists and trainees to observe surgical details repeatedly. This paper presents a deep learning model for real-time identification of surgical instruments in these videos, using a custom dataset scraped from open-access sources. Inspired by the architecture of YOLOV9, the model employs a Programmable Gradient Information (PGI) mechanism and a novel Generally-Optimized Efficient Layer Aggregation Network (Go-ELAN) to address the information bottleneck problem, enhancing Minimum Average Precision (mAP) at higher Non-Maximum Suppression Intersection over Union (NMS IoU) scores. The Go-ELAN YOLOV9 model, evaluated against YOLO v5, v7, v8, v9 vanilla, Laptool and DETR, achieves a superior mAP of 73.74 at IoU 0.5 on a dataset of 615 images with 10 instrument classes, demonstrating the effectiveness of the proposed model.
Callum Simpson, Gerard Hall, John S. Duncan
et al.
MRI-based delineation of brain tissue removed by epilepsy surgery can be challenging due to post-operative brain shift. In consequence, most studies use manual approaches which are prohibitively time-consuming for large sample sizes, require expertise, and can be prone to errors. We propose RAMPS (Resections And Masks in Preoperative Space), an automated pipeline to generate a 3D resection mask of pre-operative tissue. Our pipeline leverages existing software including FreeSurfer, SynthStrip, Sythnseg and ANTS to generate a mask in the same space as the patient's pre-operative T1 weighted MRI. We compare our automated masks against manually drawn masks and two other existing pipelines (Epic-CHOP and ResectVol). Comparing to manual masks (N=87), RAMPS achieved a median(IQR) dice similarity of 0.86(0.078) in temporal lobe resections, and 0.72(0.32) in extratemporal resections. In comparison to other pipelines, RAMPS had higher dice similarities (N=62) (RAMPS:0.86, Epic-CHOP: 0.72, ResectVol: 0.72). We release a user-friendly, easy to use pipeline, RAMPS, open source for accurate delineation of resected tissue.
Pieter Smet, Martina Doneda, Ettore Lanzarone
et al.
The availability of downstream resources plays is critical in planning the admission of elective surgery patients. The most crucial one is inpatient beds. To ensure bed availability, hospitals may use machine learning (ML) models to predict patients' length-of-stay (LOS) in the admission planning stage. However, the real value of the LOS for each patient may differ from the predicted one, potentially making the schedule infeasible. To address such infeasibilities, it is possible to implement rescheduling strategies that take advantage of operational flexibility. For example, planners may postpone admission dates, relocate patients to different wards, or even transfer patients who are already admitted among wards. A straightforward assumption is that better LOS predictions can help reduce the impact of rescheduling. However, the training process of ML models that can make such accurate predictions can be very costly. Building on previous work that proposed simulated ML for evaluating data-driven approaches, this paper explores the relationship between LOS prediction accuracy and rescheduling flexibility across various corrective policies. Specifically, we examine the most effective patient rescheduling strategies under LOS prediction errors to prevent bed overflows while optimizing resource utilization
Laparoscopic surgery constrains surgeons spatial awareness because procedures are performed through a monocular, two-dimensional (2D) endoscopic view. Conventional training methods using dry-lab models or recorded videos provide limited depth cues, often leading trainees to misjudge instrument position and perform ineffective or unsafe maneuvers. To address this limitation, we present an AI-assisted training framework developed in NVIDIA Isaac Sim that couples the standard 2D laparoscopic feed with synchronized three-dimensional (3D) visual feedback delivered through a mixed-reality (MR) interface. While trainees operate using the clinical 2D view, validated AI modules continuously localize surgical instruments and detect instrument-tissue interactions in the background. When spatial misjudgments are detected, 3D visual feedback are displayed to trainees, while preserving the original operative perspective. Our framework considers various surgical tasks including navigation, manipulation, transfer, cutting, and suturing. Visually similar 2D cases can be disambiguated through the added 3D context, improving depth perception, contact awareness, and tool orientation understanding.
Despite advances in surgical techniques and care, postoperative complications are prevalent and effects up to 15% of the patients who underwent a major surgery. The objective of this study is to develop and validate models for predicting postoperative complications and death after major surgery on a large and multicenter dataset, following the previously validated MySurgeryRisk algorithm. This retrospective, longitudinal and multicenter cohort analysis included 508,097 encounters from 366,875 adult inpatients who underwent major surgeries and were admitted to healthcare institutions within the OneFlorida+ network between 01/01/2012 and 04/29/2023. We applied the validated feature selection and transformation approach in MySurgeryRisk models and redeveloped eXtreme Gradient Boosting (XGBoost) models for predicting risk of postoperative acute kidney injury (AKI), need for intensive care unit (ICU) admission, need for mechanical ventilation (MV) therapy and in-hospital mortality on a development set and evaluated the model performance on a validation set. Area under the receiver operating characteristics curve values were obtained for need for ICU admission, 0.93 (95% Confidence Interval [CI], 0.93-0.93); need for MV, 0.94 (95% CI, 0.94-0.94); AKI, 0.92 (95% CI, 0.92-0.92); and in-hospital mortality, 0.95 (95% CI, 0.94-0.95). Area under the precision-recall curve values were computed for need for ICU admission, 0.62 (95% CI, 0.62-0.63); need for MV, 0.51 (95% CI, 0.49-0.52); AKI, 0.53 (95% CI, 0.53-0.54); and in-hospital mortality, 0.26 (95% CI, 0.24-0.29). The performance of these models is comparable to that of the previously validated MySurgeryRisk models, suggesting the enhanced generalizability of the models. Primary procedure code and provider specialty consistently appeared as the top influential variables, providing valuable insights into the factors influencing surgical outcomes.
Alexandra Thomieres, Dhruva Khanzode, Emilie Duchalais
et al.
The integration of medical imaging, computational analysis, and robotic technology has brought about a significant transformation in minimally invasive surgical procedures, particularly in the realm of laparoscopic rectal surgery (LRS). This specialized surgical technique, aimed at addressing rectal cancer, requires an in-depth comprehension of the spatial dynamics within the narrow space of the pelvis. Leveraging Magnetic Resonance Imaging (MRI) scans as a foundational dataset, this study incorporates them into Computer-Aided Design (CAD) software to generate precise three-dimensional (3D) reconstructions of the patient's anatomy. At the core of this research is the analysis of the surgical workspace, a critical aspect in the optimization of robotic interventions. Sophisticated computational algorithms process MRI data within the CAD environment, meticulously calculating the dimensions and contours of the pelvic internal regions. The outcome is a nuanced understanding of both viable and restricted zones during LRS, taking into account factors such as curvature, diameter variations, and potential obstacles. This paper delves deeply into the complexities of workspace analysis for robotic LRS, illustrating the seamless collaboration between medical imaging, CAD software, and surgical robotics. Through this interdisciplinary approach, the study aims to surpass traditional surgical methodologies, offering novel insights for a paradigm shift in optimizing robotic interventions within the complex environment of the pelvis.
Daniel Bochen Tan, Murphy Yuezhen Niu, Craig Gidney
Quantum error correction is necessary for large-scale quantum computing. A promising quantum error correcting code is the surface code. For this code, fault-tolerant quantum computing (FTQC) can be performed via lattice surgery, i.e., splitting and merging patches of code. Given the frequent use of certain lattice-surgery subroutines (LaS), it becomes crucial to optimize their design in order to minimize the overall spacetime volume of FTQC. In this study, we define the variables to represent LaS and the constraints on these variables. Leveraging this formulation, we develop a synthesizer for LaS, LaSsynth, that encodes a LaS construction problem into a SAT instance, subsequently querying SAT solvers for a solution. Starting from a baseline design, we can gradually invoke the solver with shrinking spacetime volume to derive more compact designs. Due to our foundational formulation and the use of SAT solvers, LaSsynth can exhaustively explore the design space, yielding optimal designs in volume. For example, it achieves 8% and 18% volume reduction respectively over two states-of-the-art human designs for the 15-to-1 T-factory, a bottleneck in FTQC.
Anne-Marie Rickmann, Murong Xu, Tom Nuno Wolf
et al.
The wide range of research in deep learning-based medical image segmentation pushed the boundaries in a multitude of applications. A clinically relevant problem that received less attention is the handling of scans with irregular anatomy, e.g., after organ resection. State-of-the-art segmentation models often lead to organ hallucinations, i.e., false-positive predictions of organs, which cannot be alleviated by oversampling or post-processing. Motivated by the increasing need to develop robust deep learning models, we propose HALOS for abdominal organ segmentation in MR images that handles cases after organ resection surgery. To this end, we combine missing organ classification and multi-organ segmentation tasks into a multi-task model, yielding a classification-assisted segmentation pipeline. The segmentation network learns to incorporate knowledge about organ existence via feature fusion modules. Extensive experiments on a small labeled test set and large-scale UK Biobank data demonstrate the effectiveness of our approach in terms of higher segmentation Dice scores and near-to-zero false positive prediction rate.
Objective: The computation of anatomical information and laparoscope position is a fundamental block of surgical navigation in Minimally Invasive Surgery (MIS). Recovering a dense 3D structure of surgical scene using visual cues remains a challenge, and the online laparoscopic tracking primarily relies on external sensors, which increases system complexity. Methods: Here, we propose a learning-driven framework, in which an image-guided laparoscopic localization with 3D reconstructions of complex anatomical structures is obtained. To reconstruct the 3D structure of the whole surgical environment, we first fine-tune a learning-based stereoscopic depth perception method, which is robust to the texture-less and variant soft tissues, for depth estimation. Then, we develop a dense visual reconstruction algorithm to represent the scene by surfels, estimate the laparoscope poses and fuse the depth maps into a unified reference coordinate for tissue reconstruction. To estimate poses of new laparoscope views, we achieve a coarse-to-fine localization method, which incorporates our reconstructed 3D model. Results: We evaluate the reconstruction method and the localization module on three datasets, namely, the stereo correspondence and reconstruction of endoscopic data (SCARED), the ex-vivo phantom and tissue data collected with Universal Robot (UR) and Karl Storz Laparoscope, and the in-vivo DaVinci robotic surgery dataset, where the reconstructed 3D structures have rich details of surface texture with an accuracy error under 1.71 mm and the localization module can accurately track the laparoscope with only images as input. Conclusions: Experimental results demonstrate the superior performance of the proposed method in 3D anatomy reconstruction and laparoscopic localization. Significance: The proposed framework can be potentially extended to the current surgical navigation system.
Theodoros Pissas, Claudio Ravasio, Lyndon Da Cruz
et al.
Our work proposes neural network design choices that set the state-of-the-art on a challenging public benchmark on cataract surgery, CaDIS. Our methodology achieves strong performance across three semantic segmentation tasks with increasingly granular surgical tool class sets by effectively handling class imbalance, an inherent challenge in any surgical video. We consider and evaluate two conceptually simple data oversampling methods as well as different loss functions. We show significant performance gains across network architectures and tasks especially on the rarest tool classes, thereby presenting an approach for achieving high performance when imbalanced granular datasets are considered. Our code and trained models are available at https://github.com/RViMLab/MICCAI2021_Cataract_semantic_segmentation and qualitative results on unseen surgical video can be found at https://youtu.be/twVIPUj1WZM.
A knot $K\subset S^3$ is called $SU(2)$-abundant if it satisfies two conditions: first, for all but finitely many $r\in\mathbb{Q}\backslash\{0\}$, there exists an irreducible representation $π_1(S^3_r(K))\to SU(2)$; second, any slope $r=u/v\neq 0$ for which $S^3_r(K)$ admits no irreducible $SU(2)$ representation must satisfy $Δ_K(ζ^2)= 0$ for some $u$-th root of unity $ζ$. We show that if a nontrivial knot $K\subset S^3$ is not $SU(2)$-abundant then it is a prime knot whose Alexander polynomial $Δ_K(t)$ has coefficients restricted to $\{-1,0,1\}$. This implies, in particular, that all hyperbolic alternating knots are $SU(2)$-abundant. Our proof hinges on a large surgery formula connecting instanton knot homology $KHI(S^3,K)$ and framed instanton homology $I^\sharp(S^3_n(K))$ for integers $n$ satisfying $|n|\ge 2g(K)+1$. Using this technique, we derive several interesting results in instanton Floer homology: for any Berge knot $K$, the spaces $KHI(S^3,K)$ and $\widehat{HFK}(S^3,K)$ have identical dimension; for any dual knot $K_r\subset S^3_r(K)$ of a Berge knot $K$ with $r> 2g(K)-1$, we prove $\dim_\mathbb{C}KHI(S^3_r(K),K_r)=|H_1(S^3_r(K);\mathbb{Z})|$; and for any genus-one alternating knot $K$ and any $r\in\mathbb{Q}\backslash\{0\}$, the spaces $I^\sharp(S^3_r(K))$ and $\widehat{HF}(S_r^3(K))$ have equal dimension.
Yiming Xiao, Jonathan C. Lau, Dimuthu Hemachandra
et al.
Deep brain stimulation (DBS) is an effective therapy as an alternative to pharmaceutical treatments for Parkinson's disease (PD). Aside from factors such as instrumentation, treatment plans, and surgical protocols, the success of the procedure depends heavily on the accurate placement of the electrode within the optimal therapeutic targets while avoiding vital structures that can cause surgical complications and adverse neurologic effects. While specific surgical techniques for DBS can vary, interventional guidance with medical imaging has greatly contributed to the development, outcomes, and safety of the procedure. With rapid development in novel imaging techniques, computational methods, and surgical navigation software, as well as growing insights into the disease and mechanism of action of DBS, modern image guidance is expected to further enhance the capacity and efficacy of the procedure in treating PD. This article surveys the state-of-the-art techniques in image-guided DBS surgery to treat PD, and discuss their benefits and drawbacks, as well as future directions on the topic.
Cyber-physical systems for robotic surgery have enabled minimally invasive procedures with increased precision and shorter hospitalization. However, with increasing complexity and connectivity of software and major involvement of human operators in the supervision of surgical robots, there remain significant challenges in ensuring patient safety. This paper presents a safety monitoring system that, given the knowledge of the surgical task being performed by the surgeon, can detect safety-critical events in real-time. Our approach integrates a surgical gesture classifier that infers the operational context from the time-series kinematics data of the robot with a library of erroneous gesture classifiers that given a surgical gesture can detect unsafe events. Our experiments using data from two surgical platforms show that the proposed system can detect unsafe events caused by accidental or malicious faults within an average reaction time window of 1,693 milliseconds and F1 score of 0.88 and human errors within an average reaction time window of 57 milliseconds and F1 score of 0.76.
We study the Ricci flow for initial metrics with positive isotropic curvature (strictly PIC for short). In the first part of this paper, we prove new curvature pinching estimates which ensure that blow-up limits are uniformly PIC in all dimensions. Moreover, in dimension $n \geq 12$, we show that blow-up limits are weakly PIC2. This can be viewed as a higher-dimensional version of the fundamental Hamilton-Ivey pinching estimate in dimension $3$. In the second part, we develop a theory of ancient solutions which have bounded curvature; are $κ$-noncollapsed; are weakly PIC2; and are uniformly PIC. This is an extension of Perelman's work; the additional ingredients needed in the higher dimensional setting are the differential Harnack inequality for solutions to the Ricci flow satisfying the PIC2 condition, and a rigidity result due to Brendle-Huisken-Sinestrari for ancient solutions that are uniformly PIC1. In the third part of this paper, we prove a Canonical Neighborhood Theorem for the Ricci flow with initial data with positive isotropic curvature, which holds in dimension $n \geq 12$. This relies on the curvature pinching estimates together with the structure theory for ancient solutions. This allows us to adapt Perelman's surgery procedure to this situation. As a corollary, we obtain a topological classification of all compact manifolds with positive isotropic curvature of dimension $n \geq 12$ which do not contain non-trivial incompressible $(n-1)$-dimensional space forms.
We investigate grasping of rigid objects in unilateral robot-assisted minimally invasive surgery (RAMIS) in this paper. We define a human-centered transparency that quantifies natural action and perception in RAMIS. We demonstrate this human-centered transparency analysis for different values of gripper scaling - the scaling between the grasp aperture of the surgeon-side manipulator and the aperture of the surgical instrument grasper. Thirty-one participants performed teleoperated grasping and perceptual assessment of rigid objects in one of three gripper scaling conditions (fine, normal, and quick, trading off precision and responsiveness). Psychophysical analysis of the variability of maximal grasping aperture during prehension and of the reported size of the object revealed that in normal and quick (but not in the fine) gripper scaling conditions, teleoperated grasping with our system was similar to natural grasping, and therefore, human-centered transparent. We anticipate that using motor control and psychophysics for human-centered optimizing of teleoperation control will eventually improve the usability of RAMIS.
Mahtab J. Fard, Sattar Ameri, Ratna B. Chinnam
et al.
Evaluating surgeon skill has predominantly been a subjective task. Development of objective methods for surgical skill assessment are of increased interest. Recently, with technological advances such as robotic-assisted minimally invasive surgery (RMIS), new opportunities for objective and automated assessment frameworks have arisen. In this paper, we applied machine learning methods to automatically evaluate performance of the surgeon in RMIS. Six important movement features were used in the evaluation including completion time, path length, depth perception, speed, smoothness and curvature. Different classification methods applied to discriminate expert and novice surgeons. We test our method on real surgical data for suturing task and compare the classification result with the ground truth data (obtained by manual labeling). The experimental results show that the proposed framework can classify surgical skill level with relatively high accuracy of 85.7%. This study demonstrates the ability of machine learning methods to automatically classify expert and novice surgeons using movement features for different RMIS tasks. Due to the simplicity and generalizability of the introduced classification method, it is easy to implement in existing trainers.
E. Solfaroli Camillocci, G. Baroni, F. Bellini
et al.
The background induced by the high penetration power of the gamma radiation is the main limiting factor of the current Radio-guided surgery (RGS). To partially mitigate it, a RGS with beta+ emitting radio-tracers has been suggested in literature. Here we propose the use of beta- emitting radio-tracers and beta- probes and discuss the advantage of this method with respect to the previously explored ones: the electron low penetration power allows for simple and versatile probes and could extend RGS to tumours for which background originating from nearby healthy tissue makes gamma probes less effective. We developed a beta- probe prototype and studied its performances on phantoms. By means of a detailed simulation we have also extrapolated the results to estimate the performances in a realistic case of meningioma, pathology which is going to be our first in-vivo test case. A good sensitivity to residuals down to 0.1ml can be reached within 1s with an administered activity smaller than those for PET-scans thus making the radiation exposure to medical personnel negligible.