Hasil untuk "Surgery"

S. Berríos-Torres, C. Umscheid, D. Bratzler et al.

A. Wein

P. Clavien, J. Barkun, M. D. de Oliveira et al.

D. Jayne, P. Guillou, H. Thorpe et al.

S. Nashef, F. Roques, P. Michel et al.

L. Galatz, C. Ball, S. Teefey et al.

Guozhi Liu, Weiwei Lin, Tiansheng Huang et al.

Harmful fine-tuning can invalidate safety alignment of large language models, exposing significant safety risks. In this paper, we utilize the attention sink mechanism to mitigate harmful fine-tuning. Specifically, we first measure a statistic named \emph{sink divergence} for each attention head and observe that \emph{different attention heads exhibit two different signs of sink divergence}. To understand its safety implications, we conduct experiments and find that the number of attention heads of positive sink divergence increases along with the increase of the model's harmfulness when undergoing harmful fine-tuning. Based on this finding, we propose a separable sink divergence hypothesis -- \emph{attention heads associating with learning harmful patterns during fine-tuning are separable by their sign of sink divergence}. Based on the hypothesis, we propose a fine-tuning-stage defense, dubbed Surgery. Surgery utilizes a regularizer for sink divergence suppression, which steers attention heads toward the negative sink divergence group, thereby reducing the model's tendency to learn and amplify harmful patterns. Extensive experiments demonstrate that Surgery improves defense performance by 5.90\%, 11.25\%, and 9.55\% on the BeaverTails, HarmBench, and SorryBench benchmarks, respectively. Source code is available on https://github.com/Lslland/Surgery.

Geunyoung Kim

The classical Seifert algorithm provides an explicit construction of a Seifert surface for any link in $S^3$. Alegria and Menasco extended this construction to integral homology $3$-spheres using Heegaard splittings. In this paper, we extend the Seifert algorithm to null-homologous links in arbitrary $3$-manifolds via surgery on framed links in $S^3$.

Alessandro Pisana, Barak Shoshany, Stathis Antoniou et al.

We construct a model for the nucleation of a wormhole within a Lorentzian spacetime by employing techniques from topological surgery and Morse theory. In our framework, a 0-surgery process describes the neighborhood of the nucleation point inside a compact region of spacetime, yielding a singular Lorentzian cobordism that connects two spacelike regions with different topologies. To avoid the singularity at the critical point of the Morse function, we employ the Misner trick of taking a connected sum with a closed 4-manifold -- namely $\mathbb{CP}^{2}$ -- to obtain an everywhere nondegenerate Lorentzian metric. This connected sum replaces the naked singularity with a region containing closed timelike curves. The obtained spacetime is nonsingular, but violates all the standard energy conditions. Our construction, thus, shows that a wormhole can be "created" without singularities in classical general relativity.

Leslee J Shaw, Todd C Villines, Francesco Giammarile et al.

Background Understanding pandemic-related reductions and subsequent recovery of cardiovascular testing in Asia is important for guiding regional public health efforts.Objectives This study sought to evaluate the recovery of cardiovascular testing in Asia 1 year into the COVID-19 pandemic.Methods In this subanalysis of a worldwide survey on the impact of COVID-19 on cardiovascular diagnostic care in April 2020 and April 2021, recovery of testing volume in Asia was compared among subregions, World Bank income groups and imaging modalities.Results Of 669 sites worldwide, 164 sites were in 33 Asian countries. Cardiovascular testing volumes in Asia decreased by 53% from March 2019 to April 2020, then recovered 96% of this decrease by April 2021, compared with 98% recovery in the rest of the world. Eastern Asia and Western and Central Asia reported recovery rates of 123% and 110%, compared with 50% and 80% recovery in Southern and South-eastern Asia. Testing volumes among high-income and upper-middle-income Asian countries recovered to 117% and 121% but remained depressed at 49% and 14% recovery in lower-middle and low-income countries, respectively. Stress ECG, stress echo and stress positron emission tomography studies experienced median reductions of 48%, 35% and 57% in testing volume between March 2019 and April 2021, while volumes of coronary artery calcium, coronary CT angiography and cardiac MR remained stable during this period.Conclusions The recovery of cardiovascular testing in Asia 1 year into the COVID-19 pandemic lagged in the Southern and South-eastern subregions, as well as in lower-income countries. Recovery favoured advanced cardiac imaging modalities over standard stress testing modalities.

Alberto Busetto, Giorgio Cannone, Luigi Lione et al.

ABSTRACT Chylothorax is a rare but insidious condition, characterized by the accumulation of chyle in the pleural space, which is particularly common after cardiothoracic surgeries. It presents significant challenges in both diagnosis and treatment. In this technical report, we present our experience in managing four cases of postsurgical chylothorax, each one treated with a different approach. The first and second cases were successfully managed with Lipiodol lymphangiography, which allowed for the visualization and occlusion of the injured lymphatic duct, leading to the resolution of the chylothorax. The third case involved thoracic duct embolization, a procedure that resulted in the closure of the duct responsible for the chylous effusion. The last case involved a patient who developed left‐sided chylothorax following a pulmonary resection. The patient experienced chylous leakage early in the postoperative period and underwent a revision thoracoscopy for hemostasis and thoracic duct ligation. During the procedure, indocyanine green (ICG) fluorescence was used to effectively identify and ligate the injured chylous duct. This case series highlights the variety of therapeutic strategies available for the management of chylothorax, emphasizing the importance of a structured, stepwise approach tailored to the specific needs of each patient.

Qing-Quan Chen, Qing-Quan Chen, Han-Lin Chen et al.

BackgroundThe incidence of secondary displacement in fractures of the greater tuberosity of the humerus remains high, irrespective of whether conservative or surgical treatment is administered. However, the specific risk factors contributing to secondary displacement of the greater tuberosity of the humerus have not been previously reported. The primary objective of this study was to analyze the risk factors associated with secondary displacement of the greater tuberosity of the humerus and to summarize corresponding guidelines for clinical diagnosis and treatment.MethodsA retrospective analysis was conducted on patients with fractures of the greater tuberosity of the humerus who received treatment at the same trauma center between January 2018 and December 2022. The following variables were recorded for each patient: age, gender, injured limb (left/right), whether the fracture was comminuted, bone density, fracture displacement, shoulder joint dislocation, treatment plan, and treatment outcomes, including the success rate of reduction and the time of secondary displacement. The patients were categorized into two groups based on the absence or presence of secondary displacement. For statistical analysis, the Mann–Whitney U test and logistic regression analysis were employed. The significance level was set at P < 0.05.ResultsAmong the 177 patients enrolled in this study, 144 (81.36%) did not exhibit secondary displacement, while 33 (18.64%) did present with such displacement. Significant statistical differences were observed between the two groups in mean age, fracture type, bone mineral density, shoulder dislocation, and reduction quality of fracture, indicating a statistically significant association (P < 0.05). However, no significant difference was found in gender, Left/right limbs, displacement of fracture, and treatment method (P > 0.05). Logistic regression analysis revealed that comminuted fractures, osteoporosis, shoulder dislocation and poor reduction independently contributed to an increased risk of secondary displacement of the greater tuberosity of humerus.ConclusionsComminuted fracture, osteoporosis, shoulder dislocation, and poor reduction have been identified as independent risk factors for secondary displacement. In the course of clinical diagnosis and treatment, it is imperative to consider the potential adverse prognosis that may be associated with these conditions.

Siddhart Yadav, K P Chiranjeevi, Akash Singh Jadon et al.

Introduction: Ipsilateral tibia and fibula shaft fractures with trimalleolar fracture are quite rare in clinical practice. Case Report: This is a case report of a 49-year-old female presented on March 6th, 2024, and was diagnosed to have an ipsilateral left comminuted distal tibia shaft and fibula shaft fracture with an anterior lacerated wound 2 cm over the fracture site with trimalleolar fracture after falling twice while walking. The patient was treated with wound debridement, intramedullary interlocking nailing for the left tibia shaft, and open reduction internal fixation with coracoclavicular screw for posterior malleolus, K-wires + FiberWire tension band wiring for medial malleolus, and K-wires for lateral malleolus on March 07th, 2024. K-wires from the lateral malleolus were removed and tibia nail dynamization was done on April 10th, 2024. All fractures united in 4 months and the patient was followed up for a period of 1 year post-operatively. Conclusion: Various treatment options were possible, of which we chose implants and a sequence of fixation based on the fracture pattern being comminuted and an open fracture.

Beilei Cui, Mobarakol Islam, Long Bai et al.

Purpose: Depth estimation in robotic surgery is vital in 3D reconstruction, surgical navigation and augmented reality visualization. Although the foundation model exhibits outstanding performance in many vision tasks, including depth estimation (e.g., DINOv2), recent works observed its limitations in medical and surgical domain-specific applications. This work presents a low-ranked adaptation (LoRA) of the foundation model for surgical depth estimation. Methods: We design a foundation model-based depth estimation method, referred to as Surgical-DINO, a low-rank adaptation of the DINOv2 for depth estimation in endoscopic surgery. We build LoRA layers and integrate them into DINO to adapt with surgery-specific domain knowledge instead of conventional fine-tuning. During training, we freeze the DINO image encoder, which shows excellent visual representation capacity, and only optimize the LoRA layers and depth decoder to integrate features from the surgical scene. Results: Our model is extensively validated on a MICCAI challenge dataset of SCARED, which is collected from da Vinci Xi endoscope surgery. We empirically show that Surgical-DINO significantly outperforms all the state-of-the-art models in endoscopic depth estimation tasks. The analysis with ablation studies has shown evidence of the remarkable effect of our LoRA layers and adaptation. Conclusion: Surgical-DINO shed some light on the successful adaptation of the foundation models into the surgical domain for depth estimation. There is clear evidence in the results that zero-shot prediction on pre-trained weights in computer vision datasets or naive fine-tuning is not sufficient to use the foundation model in the surgical domain directly. Code is available at https://github.com/BeileiCui/SurgicalDINO.

Jingying Wang, Haoran Tang, Taylor Kantor et al.

Videos are prominent learning materials to prepare surgical trainees before they enter the operating room (OR). In this work, we explore techniques to enrich the video-based surgery learning experience. We propose Surgment, a system that helps expert surgeons create exercises with feedback based on surgery recordings. Surgment is powered by a few-shot-learning-based pipeline (SegGPT+SAM) to segment surgery scenes, achieving an accuracy of 92\%. The segmentation pipeline enables functionalities to create visual questions and feedback desired by surgeons from a formative study. Surgment enables surgeons to 1) retrieve frames of interest through sketches, and 2) design exercises that target specific anatomical components and offer visual feedback. In an evaluation study with 11 surgeons, participants applauded the search-by-sketch approach for identifying frames of interest and found the resulting image-based questions and feedback to be of high educational value.

Peng Xiao, Xuechun Cai, Zhou Zhang et al.

Abstract Most patients with inflammatory bowel disease (IBD) develop anemia, which is attributed to the dysregulation of iron metabolism. Reciprocally, impaired iron homeostasis also aggravates inflammation. How this iron‐mediated, pathogenic anemia‐inflammation crosstalk is regulated in the gut remains elusive. Herein, it is for the first time revealed that anemic IBD patients exhibit impaired production of short‐chain fatty acids (SCFAs), particularly butyrate. Butyrate supplementation restores iron metabolism in multiple anemia models. Mechanistically, butyrate upregulates ferroportin (FPN) expression in macrophages by reducing the enrichment of histone deacetylase (HDAC) at the Slc40a1 promoter, thereby facilitating iron export. By preventing iron sequestration, butyrate not only mitigates colitis‐induced anemia but also reduces TNF‐α production in macrophages. Consistently, macrophage‐conditional FPN knockout mice exhibit more severe anemia and inflammation. Finally, it is revealed that macrophage iron overload impairs the therapeutic effectiveness of anti‐TNF‐α antibodies in colitis, which can be reversed by butyrate supplementation. Hence, this study uncovers the pivotal role of butyrate in preventing the pathogenic circuit between anemia and inflammation.

Tyler LeBlond, Christopher Dean, George Watkins et al.

We report a resource estimation pipeline that explicitly compiles quantum circuits expressed using the Clifford+T gate set into a surface code lattice surgery instruction set. The cadence of magic state requests from the compiled circuit enables the optimization of magic state distillation and storage requirements in a post-hoc analysis. To compile logical circuits into lattice surgery operations, we build upon the open-source Lattice Surgery Compiler. The revised compiler operates in two stages: the first translates logical gates into an abstract, layout-independent instruction set; the second compiles these into local lattice surgery instructions that are allocated to hardware tiles according to a specified resource layout. The second stage retains logical parallelism while avoiding resource contention in the fault-tolerant layer, aiding realism. Additionally, users can specify dedicated tiles at which magic states are replenished, enabling resource costs from the logical computation to be considered independently from magic state distillation and storage. We demonstrate the applicability of our pipeline to large, practical quantum circuits by providing resource estimates for the ground state estimation of molecules. We find that variable magic state consumption rates in real circuits can cause the resource costs of magic state storage to dominate unless production is varied to suit.

Xiang Yang, Dezhi Chen, Qi Qi et al.

Deep Neural Networks (DNNs) have significantly improved the accuracy of intelligent applications on mobile devices. DNN surgery, which partitions DNN processing between mobile devices and multi-access edge computing (MEC) servers, can enable real-time inference despite the computational limitations of mobile devices. However, DNN surgery faces a critical challenge: determining the optimal computing resource demand from the server and the corresponding partition strategy, while considering both inference latency and MEC server usage costs. This problem is compounded by two factors: (1) the finite computing capacity of the MEC server, which is shared among multiple devices, leading to inter-dependent demands, and (2) the shift in modern DNN architecture from chains to directed acyclic graphs (DAGs), which complicates potential solutions. In this paper, we introduce a novel Decentralized DNN Surgery (DDS) framework. We formulate the partition strategy as a min-cut and propose a resource allocation game to adaptively schedule the demands of mobile devices in an MEC environment. We prove the existence of a Nash Equilibrium (NE), and develop an iterative algorithm to efficiently reach the NE for each device. Our extensive experiments demonstrate that DDS can effectively handle varying MEC scenarios, achieving up to 1.25$\times$ acceleration compared to the state-of-the-art algorithm.