Laparoscopic Video Research Articles

A comparative evaluation of spatial pooling methods for surgical tool detection

Abstract Surgical tool detection is an important aspect for recognising surgical activities and understanding surgical workflow. Laparoscopic videos represent an information source that can be used for recognising surgical tools. However, manual labelling of tool incidence and location in such data is extremely time intensive. Therefore, weaklysupervised approaches have been developed to perform tool localisation. In this study, three types of spatial pooling methods were implemented to evaluate the influence of each method on the performance of weakly-supervised model. The best achieved performance was a mean average precision (mAP) of 94% for tool classification and a f1-score of 70% for tool localisation. Experimental results showed the importance of selecting an appropriate pooling function to enhance model performance.

Development and validation of GLVS (Generic Laparoscopic Video Scoring System), a tool for assessment of laparoscopic skills in gynaecology using videotaped procedures: Are edited videos representative of their full-length version?

Background: Anonymized videotaped endoscopic procedures can be used for the assessment of surgical competence, but a reliable non-procedure-specific scoring system is needed for gynaecology. Objectives: To design and evaluate the validity of the Generic Laparoscopic Video Scoring System (GLVS), a novel tool in the assessment of various gynaecological laparoscopic procedures. Materials and methods: Seventeen anonymized unedited video recordings of various gynaecological laparoscopic procedures and the 4-minute-long edited versions of the same videos were independently scored by two experts, twice, using GLVS. Main outcome measures: Internal consistency reliability, test-retest, and inter-rater reliability of GLVS. We also compared the scored achieved by edited videos with those of the full-length version of the same videos. Results: The mean score achieved by 4-minute-long edited videos was similar to that of the unedited version (p= 0.13 - 0.19). There was excellent correlation between the pooled scores for edited and unedited versions (intra-class correlation coefficient = 0.86). GLVS had excellent internal consistency reliability (Cronbach’s alpha 0.92-0.97). Test-retest and inter-rater reliability were generally better for edited 4-minute-long videos compared to their full-length version. Test-retest reliability for edited videos was excellent for scorer 1 and good for scorer 2 with intra-class correlation coefficient (ICC) of 0.88 and 0.62 respectively. Inter-rater reliability was good for edited videos (ICC=0.64) but poor for full-length versions (ICC= -0.24). Conclusion: GLVS allows for objective surgical skills assessment using anonymized shortened self-edited videos of basic gynaecological laparoscopic procedures. Shortened video clips of procedures seem to be representative of their full-length version for the assessment of surgical skills. What’s new? We devised and undertook a validation study for a novel tool to assess surgical skills using surgical video clips. We believe this addition clearly delineates the unique contributions of our study.

Inpainting surgical occlusion from laparoscopic video sequences for robot-assisted interventions.

Medical technology for minimally invasive surgery has undergone a paradigm shift with the introduction of robot-assisted surgery. However, it is very difficult to track the position of the surgical tools in a surgical scene, so it is crucial to accurately detect and identify surgical tools. This task can be aided by deep learning-based semantic segmentation of surgical video frames. Furthermore, due to the limited working and viewing areas of these surgical instruments, there is a higher chance of complications from tissue injuries (e.g., tissue scars and tears). With the aid of digital inpainting algorithms, we present an application that uses image segmentation to remove surgical instruments from laparoscopic/endoscopic video. We employ a modified U-Net architecture (U-NetPlus) to segment the surgical instruments. It consists of a redesigned decoder and a pre-trained VGG11 or VGG16 encoder. The decoder was modified by substituting an up-sampling operation based on nearest-neighbor interpolation for the transposed convolution operation. Furthermore, these interpolation weights do not need to be learned to perform upsampling, which eliminates the artifacts generated by the transposed convolution. In addition, we use a very fast and adaptable data augmentation technique to further enhance performance. The instrument segmentation mask is filled in (i.e., inpainted) by the tool removal algorithms using the previously acquired tool segmentation masks and either previous instrument-containing frames or instrument-free reference frames. We have shown the effectiveness of the proposed surgical tool segmentation/removal algorithms on a robotic instrument dataset from the MICCAI 2015 and 2017 EndoVis Challenge. We report a 90.20% DICE for binary segmentation, a 76.26% DICE for instrument part segmentation, and a 46.07% DICE for instrument type (i.e., all instruments) segmentation on the MICCAI 2017 challenge dataset using our U-NetPlus architecture, outperforming the results of earlier techniques used and tested on these data. In addition, we demonstrated the successful execution of the tool removal algorithm from surgical tool-free videos that contained moving surgical tools that were generated artificially. Our application successfully separates and eliminates the surgical tool to reveal a view of the background tissue that was otherwise hidden by the tool, producing results that are visually similar to the actual data.

Development of a deep learning model for safe direct optical trocar insertion in minimally invasive surgery: an innovative method to prevent trocar injuries.

Direct optical trocar insertion is a common procedure in laparoscopic minimally invasive surgery. However, misinterpretations of the abdominal wall anatomy can lead to severe complications. Artificial intelligence has shown promise in surgical endoscopy, particularly in the employment of deep learning models for anatomical landmark identification. This study aimed to integrate a deep learning model with an alarm system algorithm for the precise detection of abdominal wall layers during trocar placement. Annotated bounding boxes and assigned classes were based on the six layers of the abdominal wall: subcutaneous, anterior rectus sheath, rectus muscle, posterior rectus sheath, peritoneum, and abdominal cavity. The cutting-edge YOLOv8 model was combined with a deep learning detector to train the dataset. The model was trained on still images and inferenced on laparoscopic videos to ensure real-time detection in the operating room. The alarm system was activated upon recognizing the peritoneum and abdominal cavity layers. We assessed the model's performance using mean average precision (mAP), precision, and recall metrics. A total of 3600 images were captured from 89 laparoscopic video cases. The proposed model was trained on 3000 images, validated with a set of 200 images, and tested on a separate set of 400 images. The results from the test set were 95.8% mAP, 89.8% precision, and 91.7% recall. The alarm system was validated and accepted by experienced surgeons at our institute. We demonstrated that deep learning has the potential to assist surgeons during direct optical trocar insertion. During trocar insertion, the proposed model promptly detects precise landmark references in real-time. The integration of this model with the alarm system enables timely reminders for surgeons to tilt the scope accordingly. Consequently, the implementation of the framework provides the potential to mitigate complications associated with direct optical trocar placement, thereby enhancing surgical safety and outcomes.

Live laparoscopic video retrieval with compressed uncertainty.

Searching through large volumes of medical data to retrieve relevant information is a challenging yet crucial task for clinical care. However the primitive and most common approach to retrieval, involving text in the form of keywords, is severely limited when dealing with complex media formats. Content-based retrieval offers a way to overcome this limitation, by using rich media as the query itself. Surgical video-to-video retrieval in particular is a new and largely unexplored research problem with high clinical value, especially in the real-time case: using real-time video hashing, search can be achieved directly inside of the operating room. Indeed, the process of hashing converts large data entries into compact binary arrays or hashes, enabling large-scale search operations at a very fast rate. However, due to fluctuations over the course of a video, not all bits in a given hash are equally reliable. In this work, we propose a method capable of mitigating this uncertainty while maintaining a light computational footprint. We present superior retrieval results (3%-4% top 10 mean average precision) on a multi-task evaluation protocol for surgery, using cholecystectomy phases, bypass phases, and coming from an entirely new dataset introduced here, surgical events across six different surgery types. Success on this multi-task benchmark shows the generalizability of our approach for surgical video retrieval.

The Importance of Early Diagnosis and Treatment of Incidental Tension Pneumothorax During Robotic Assisted Video laparoscopy for Treatment of Diaphragmatic Endometriosis: Report of Two Cases

We describe two cases of robotic assisted video laparoscopy for the treatment of diaphragmatic endometriosis, in which incidental Tension pneumothorax occurred during the initial Inspection and inventory of diaphragmatic lesions. We Demonstrate the importance of early diagnosis of this complication and report successful resolution using the thoracic drainage technique.

Відеолапароскопія при закритій травмі живота з пошкодженням печінки та селезінки

Open and closed abdominal trauma is the structure of traumatic injuries makes up from 1.5% to 36.5% of all injuries, accompanied in 26.7-40.8% of cases by liver and spleen damage. Damage to parenchymal organs of the abdominal cavity most often leads to the development of intra-abdominal bleeding, which is the basis of unsatisfactory results of treatment and death of patients. Among the diagnostic methods, video laparoscopy is becoming more and more important in finding the extent of damage to the internal organs of the abdominal cavity. In emergency cases as a minimally invasive but sufficiently informative research method, which in 20-60% of cases allows to supplement, clarify or change the diagnosis. The sensitivity of diagnostic laparoscopy in the case of damage to the organs of the abdominal cavity is 99-100%. Purpose - to study the role and place of video laparoscopy in the diagnosis and treatment of closed abdominal trauma with damage to the liver and spleen. Materials and methods. The results of using video laparoscopy in the diagnosis and treatment of closed abdominal trauma with isolated liver and spleen damage in 131 patients were retrospectively analyzed. The diagnostic search program included clinical, laboratory-biochemical, radiographic studies, ultrasound examination of the abdominal organs according to the FAST protocol, computed tomography, laparocentesis using a search catheter, diagnostic peritoneal lavage, laparoscopy, which were performed in a comprehensive examination according to indications. The obtained data are presented in the form of absolute (number of cases) and relative (%) values. The validity of laparoscopy was considered statistically significant at p˂0.05. Results. Video laparoscopy of the abdominal cavity in the case of a closed injury with isolated damage to parenchymal organs allows to diagnose the nature of the damage in the shortest possible time and to determine the treatment tactics. This method in the diagnosis of hemoperitoneum and injuries of the liver and spleen is informative (100%), sensitivity - 92.8%, specificity - 94.6%, at p<0.05. All laparoscopic operations for injuries of the liver and spleen were ended with a thorough revision, sanitation and drainage of the abdominal cavity. We did not observe complications related to endo-video surgery in this group of patients. There were no fatal consequences. This indicates that the use of minimally invasive laparoscopic methods of treatment in this group of patients helps to reduce the severity of surgical trauma and the frequency of various postoperative complications. Conclusions. Laparoscopy is a highly informative method of diagnosing intra-abdominal damage to the liver and spleen in case of closed abdominal trauma with the possibility of carrying out medical measures. It allows you to prevent and significantly reduce the frequency of postoperative complications, improve the results of treatment, and shorten the hospitalization period of the patients. In compliance of the above criteria, laparoscopy is a safe and adequate replacement for laparotomy as both a diagnostic and treatment intervention. The research was carried out in accordance with the principles of the Helsinki Declaration. The study protocol was approved by the Local Ethics Committee of the participating institution. The informed consent of the patient was obtained for conducting the studies. No conflict of interests was declared by the authors.

Laparoscopic Hepp Couinaud for the treatment of Caroli's disease: A case report

Caroli disease is a rare congenital disorder characterized by dilatation of intrahepatic bile ducts. The treatment could correspond to hepatic resection for localized disease.
 This paper describes a case of Caroli’s disease in a 60 years old female patient, who presented jaundice and diffuse abdominal pain. Further examinations suggested the diagnosis of Caroli disease associated with cholestatic syndrome. We decided to perform a Hepp Couinaud biliodigestive bypass by video laparoscopy with Roux-en-Y reconstruction. There were no immediate complications and no longterm complications after 6 months of follow-up. The VLP biliodigestive bypass in patients with associated cholestatic syndromes should be considered a treatment prior to a more aggressive approach.

Successful treatment of pelvic venous incompetence by a single-stage combined laparoscopic intervention: A case report

Introduction: Pelvic venous incompetence (PVI) is the most common chronic pathology, mainly affecting women of different ages. Its characteristic signs include pelvic venous fullness syndrome and chronic pelvic pain, contributing to temporary disability and impaired reproductive function. While there are known surgical treatment methods for pelvic pain in PVI, they have numerous complications. Case presentation: A 39-year-old female patient presented with PVI and chronic pelvic pain, dysmenorrhea, and infertility. They underwent simultaneous combined reconstructiveplastic conservative surgical treatment under 3D laparoscopic video monitoring. Conclusion: The patient's treatment outcome was characterized by decreased pelvic pain, normalized menstrual function, and a long-awaited uterine pregnancy ending with the birth of a full-term baby. The two successively performed surgical treatment procedures had a cumulative therapeutic effect, relieving the patient's varicose pelvic pain due to various causes and restoring their reproductive function.

Ensuring privacy protection in the era of big laparoscopic video data: development and validation of an inside outside discrimination algorithm (IODA)

BackgroundLaparoscopic videos are increasingly being used for surgical artificial intelligence (AI) and big data analysis. The purpose of this study was to ensure data privacy in video recordings of laparoscopic surgery by censoring extraabdominal parts. An inside-outside-discrimination algorithm (IODA) was developed to ensure privacy protection while maximizing the remaining video data.MethodsIODAs neural network architecture was based on a pretrained AlexNet augmented with a long-short-term-memory. The data set for algorithm training and testing contained a total of 100 laparoscopic surgery videos of 23 different operations with a total video length of 207 h (124 min ± 100 min per video) resulting in 18,507,217 frames (185,965 ± 149,718 frames per video). Each video frame was tagged either as abdominal cavity, trocar, operation site, outside for cleaning, or translucent trocar. For algorithm testing, a stratified fivefold cross-validation was used.ResultsThe distribution of annotated classes were abdominal cavity 81.39%, trocar 1.39%, outside operation site 16.07%, outside for cleaning 1.08%, and translucent trocar 0.07%. Algorithm training on binary or all five classes showed similar excellent results for classifying outside frames with a mean F1-score of 0.96 ± 0.01 and 0.97 ± 0.01, sensitivity of 0.97 ± 0.02 and 0.0.97 ± 0.01, and a false positive rate of 0.99 ± 0.01 and 0.99 ± 0.01, respectively.ConclusionIODA is able to discriminate between inside and outside with a high certainty. In particular, only a few outside frames are misclassified as inside and therefore at risk for privacy breach. The anonymized videos can be used for multi-centric development of surgical AI, quality management or educational purposes. In contrast to expensive commercial solutions, IODA is made open source and can be improved by the scientific community.

CholecTriplet2021: A benchmark challenge for surgical action triplet recognition.

Context-aware decision support in the operating room can foster surgical safety and efficiency by leveraging real-time feedback from surgical workflow analysis. Most existing works recognize surgical activities at a coarse-grained level, such as phases, steps or events, leaving out fine-grained interaction details about the surgical activity; yet those are needed for more helpful AI assistance in the operating room. Recognizing surgical actions as triplets of ‹instrument, verb, target› combination delivers more comprehensive details about the activities taking place in surgical videos. This paper presents CholecTriplet2021: an endoscopic vision challenge organized at MICCAI 2021 for the recognition of surgical action triplets in laparoscopic videos. The challenge granted private access to the large-scale CholecT50 dataset, which is annotated with action triplet information. In this paper, we present the challenge setup and the assessment of the state-of-the-art deep learning methods proposed by the participants during the challenge. A total of 4 baseline methods from the challenge organizers and 19 new deep learning algorithms from the competing teams are presented to recognize surgical action triplets directly from surgical videos, achieving mean average precision (mAP) ranging from 4.2% to 38.1%. This study also analyzes the significance of the results obtained by the presented approaches, performs a thorough methodological comparison between them, in-depth result analysis, and proposes a novel ensemble method for enhanced recognition. Our analysis shows that surgical workflow analysis is not yet solved, and also highlights interesting directions for future research on fine-grained surgical activity recognition which is of utmost importance for the development of AI in surgery.

Beyond rankings: Learning (more) from algorithm validation.

Challenges have become the state-of-the-art approach to benchmark image analysis algorithms in a comparative manner. While the validation on identical data sets was a great step forward, results analysis is often restricted to pure ranking tables, leaving relevant questions unanswered. Specifically, little effort has been put into the systematic investigation on what characterizes images in which state-of-the-art algorithms fail. To address this gap in the literature, we (1) present a statistical framework for learning from challenges and (2) instantiate it for the specific task of instrument instance segmentation in laparoscopic videos. Our framework relies on the semantic meta data annotation of images, which serves as foundation for a General Linear Mixed Models (GLMM) analysis. Based on 51,542 meta data annotations performed on 2,728 images, we applied our approach to the results of the Robust Medical Instrument Segmentation Challenge (ROBUST-MIS) challenge 2019 and revealed underexposure, motion and occlusion of instruments as well as the presence of smoke or other objects in the background as major sources of algorithm failure. Our subsequent method development, tailored to the specific remaining issues, yielded a deep learning model with state-of-the-art overall performance and specific strengths in the processing of images in which previous methods tended to fail. Due to the objectivity and generic applicability of our approach, it could become a valuable tool for validation in the field of medical image analysis and beyond.

Surface deformation tracking in monocular laparoscopic video.

Image-guided surgery has been proven to enhance the accuracy and safety of minimally invasive surgery (MIS). Nonrigid deformation tracking of soft tissue is one of the main challenges in image-guided MIS owing to the existence of tissue deformation, homogeneous texture, smoke and instrument occlusion, etc. In this paper, we proposed a piecewise affine deformation model-based nonrigid deformation tracking method. A Markov random field based mask generation method is developed to eliminate tracking anomalies. The deformation information vanishes when the regular constraint is invalid, which further deteriorates the tracking accuracy. Atime-series deformation solidification mechanism is introduced to reduce the degradation of the deformation field of the model. For the quantitative evaluation of the proposed method, we synthesized nine laparoscopic videos mimicking instrument occlusion and tissue deformation. Quantitative tracking robustness was evaluated on the synthetic videos. Three real videos of MIS containing challenges of large-scale deformation, large-range smoke, instrument occlusion, and permanent changes in soft tissue texture were also used to evaluate the performance of the proposed method. Experimental results indicate the proposed method outperforms state-of-the-art methods in terms of accuracy and robustness, which shows good performance in image-guided MIS.

Learning feature descriptors for pre- and intra-operative point cloud matching for laparoscopic liver registration.

In laparoscopic liver surgery, preoperative information can be overlaid onto the intra-operative scene by registering a 3D preoperative model to the intra-operative partial surface reconstructed from the laparoscopic video. To assist with this task, we explore the use of learning-based feature descriptors, which, to our best knowledge, have not been explored for use in laparoscopic liver registration. Furthermore, a dataset to train and evaluate the use of learning-based descriptors does not exist. We present the LiverMatch dataset consisting of 16 preoperative models and their simulated intra-operative 3D surfaces. We also propose the LiverMatch network designed for this task, which outputs per-point feature descriptors, visibility scores, and matched points. We compare the proposed LiverMatch network with a network closest to LiverMatch and a histogram-based 3D descriptor on the testing split of the LiverMatch dataset, which includes two unseen preoperative models and 1400 intra-operative surfaces. Results suggest that our LiverMatch network can predict more accurate and dense matches than the other two methods and can be seamlessly integrated with a RANSAC-ICP-based registration algorithm to achieve an accurate initial alignment. The use of learning-based feature descriptors in laparoscopic liver registration (LLR) is promising, as it can help achieve an accurate initial rigid alignment, which, in turn, serves as an initialization for subsequent non-rigid registration.

SURGICAL TREATMENT OF VICTIMS WITH MINE AND EXPLOSIVE INJURIES DURING MILITARY ACTIONS

Summary. Mine-explosive injury is a gunshot combined multiple injury that occurs in a person as a result of the impact of the impact factors of the explosion of engineering, mine ammunition and is accompanied by deep and widespread tissue damage and contusion.
 The purpose of the work is to analyze the methods of surgical treatment of patients with mine-explosive injuries in the conditions of martial law.
 Materials and methods. The results of surgical treatment of 64 patients with mine-explosive injuries were analyzed, 27 (42.2 %) were women, 37 (57.8 %) were men, who were being treated at Dergachiv Central Hospital. All victims underwent surgical interventions depending on the location of the injury, its nature, the severity of the patient, the development of traumatic or hemorrhagic shock.
 Results and their discussion. Given that the most frequent type of injuries were wounds of various localization, the most common type of surgical intervention was surgical treatment of wounds. The scope of the intervention depended on the depth and size of the wound, as well as the nature of tissue damage. Surgical manipulations on the chest were performed to eliminate pneumothorax (pneumohydrothorax), pleural puncture was performed or pleural drainage was established. If the Rouvilois-Gregoir test was positive, a thoracotomy was performed followed by suturing of the lung wound. Invasive diagnostic methods, laparocentesis and laparoscopy were performed for suspicion of damage to the organs of the abdominal cavity, with the help of which damage to the organs of the abdominal cavity was excluded. The use of laparocentesis in such situations, although its diagnostic value is lower than that of video laparoscopy, allows to determine the pathological content in the abdominal cavity, which is sufficient to detect life-threatening consequences of abdominal injuries, for example, such as intra-abdominal bleeding.
 Conclusions. Victims with explosive lesions require a large number of surgical interventions in various anatomical areas. When providing assistance to victims of an explosive injury, one should be guided by the principles of “damage control”, determining the amount of interventions that should be performed based on vital indications.

Laparoscopic Video Analysis Using Temporal, Attention, and Multi-Feature Fusion Based-Approaches.

Adapting intelligent context-aware systems (CAS) to future operating rooms (OR) aims to improve situational awareness and provide surgical decision support systems to medical teams. CAS analyzes data streams from available devices during surgery and communicates real-time knowledge to clinicians. Indeed, recent advances in computer vision and machine learning, particularly deep learning, paved the way for extensive research to develop CAS. In this work, a deep learning approach for analyzing laparoscopic videos for surgical phase recognition, tool classification, and weakly-supervised tool localization in laparoscopic videos was proposed. The ResNet-50 convolutional neural network (CNN) architecture was adapted by adding attention modules and fusing features from multiple stages to generate better-focused, generalized, and well-representative features. Then, a multi-map convolutional layer followed by tool-wise and spatial pooling operations was utilized to perform tool localization and generate tool presence confidences. Finally, the long short-term memory (LSTM) network was employed to model temporal information and perform tool classification and phase recognition. The proposed approach was evaluated on the Cholec80 dataset. The experimental results (i.e., 88.5% and 89.0% mean precision and recall for phase recognition, respectively, 95.6% mean average precision for tool presence detection, and a 70.1% F1-score for tool localization) demonstrated the ability of the model to learn discriminative features for all tasks. The performances revealed the importance of integrating attention modules and multi-stage feature fusion for more robust and precise detection of surgical phases and tools.

Risk factors related to delayed post-anesthetic awakening in the elderly operated by elective video laparoscopic surgery

In all surgical procedures, the organism presents different physiological alterations that expose the patient to a series of difficulties that, if not taken into account, can cause very severe complications in the post-anesthetic awakening of elderly patients. A longitudinal prospective cohort study was carried out on elderly people who underwent laparoscopic video surgery at the general provincial hospital "Carlos Manuel de Céspedes de Bayamo, from the Cauto region during the period from January 1, 2019 (,) to December 31, 2019. December 2021; with the aim to identify the risk factors related to a delay in post- anesthetic awakening. It was established that the exposed cohort consisted of 39 elderly people with delayed post-anesthetic awakening in the study period who met the inclusion criteria. The magnitude of the associations was estimated by calculating the relative risks (RR) of delay. There is an association between patient-dependent risk factors: smoking, obesity, presence of comorbidities and use of chronic pharmacological therapy, with a delay in post-anesthetic awakening in the elderly. The duration of the intervention, fasting time greater than 12 hours and hypoglycemia were the factors dependent on the surgical-anesthetic act associated with a delay in post-anesthetic awakening in the elderly.

Does 3D laparoscopic video technology affect long-term survival in right hemicolectomy for cancer compared to standard 2D? A propensity score study.

There are few studies focused on the short-term results of laparoscopic right hemicolectomy performed with 2D (two-dimension) or 3D (three-dimension) video technology and none on the oncologic effects. The aim of the study was to assess the long-term results of laparoscopic right hemicolectomy (LRH) with intracorporeal anastomosis using 3D or 2D video in patients with right colon cancer with at least three years of oncologic follow-up. Data from patients undergoing laparoscopic right hemicolectomy (LRH) with intracorporeal anastomosis for cancer in an 11-year period (June 2008-June 2019) and ≥ 3years of follow-up were prospectively collected. Surgical procedures were performed by two expert laparoscopic surgeons. 111 patients were included in the study: 56 (50.5%) in the 3D group and 55 (49.5%) in the 2D group. Tumor stage and number of lymph nodes harvested were similar. Overall and disease-free survival were not different in the two groups. Local recurrence occurred in none of the patients, and distant metachronous metastases were similar in the two groups. A propensity score weighting approach was used to account for potential confounding related to patients' nonrandom allocation to the 2 groups. The effects of the intervention on postoperative outcomes were assessed with a weighted regression approach. Laparoscopic 3D technology allows similar oncological results as 2D vision in LRH with intracorporeal anastomosis. Larger prospective randomized studies might confirm these results in the long-term follow-up.

Surgical Tool Classification & Localisation Using Attention and Multi-feature Fusion Deep Learning Approach

Analysing laparoscopic videos, particularly for surgical tool classification and localisation, has attained interest in the field of surgical data science since they represent an extensive information source. However, the difficulties for acquiring and labelling these videos have led to paucity of labelled datasets. Consequently, the progress of developing robust and generalised surgical tool detection models was slowed down, and translating these models into the medical field was hindered. In this work, supervised surgical tool classification and weakly-supervised tool localisation in laparoscopic videos were addressed. A base convolutional neural network (CNN) model was adapted to perform both tasks by incorporating multi-map localisation layers. Squeeze-and-excitation modules were added to the CNN to enhance the ability of the model to generate better focused informative features. Additionally, features at multiple stages of the CNN were combined and fused in a batch normalisation layer to enhance model generalisability. The proposed model was evaluated on the popular Cholec80 dataset. Experimental results of 94.1% mean average precision for tool classification and 70.1% F1-score for tool localisation revealed the ability of the model to learn better features for both tasks. The proposed approach showed the advantages of integrating attentions modules and multi-stage features fusion technique for surgical tool classification and localisation.

Organ–preserving operations for the gun–shot thoraco–abdominal and abdominal woundings with splenic injuries

Objective.To determine the expediency of the spleen–preserving operations performance for its injury due to gun–shot thoraco–abdominal and abdominal woundings.
 Materials and methods. Of 51wounded persons with the gun–shot penetrating abdominal woundings with splenic injuries 8 (15.7%) were operated on the II level of the medical help delivery, while on the ІІІ level – 43 (84.3%). Of 43 wounded persons, operated on the ІІІ level of the medical help delivery, in 35 (81.4%) splenectomy was performed, while in 8 (18.6%) – the organ–preserving operations. All the splenic injuries have been characterized as a shrapnel woundings. The splenic damage degree was estimated in accordance to Moore, taking into account its structure. In 5 (62.5%) wounded persons the open laparotomy organ–preserving operations were performed, and while using video–laparoscopic system– in 3 (37.5%).
 Results. Ultrasonographic investigation in accordance to the FAST–protocol was done in all the wounded persons, what have permitted to determine the blood abdominal outpouring volume as well as the splenic injury presence. The affection volume and further tactics were determined using video–laparoscopy. The Stage I splenic affection in accordance to Moore was diagnosed in 4 (50%) wounded persons, the Stage ІІ – in 3 (37.5%), and the Stage ІІІ – in 1 (12.5%). In all the patients, in whom organ–preserving interventions were performed, the surgical help volume consisted of coagulation, using the radiofrequency therapy apparatus MEDTRONICAEX and the hemostatic sponge application with drainage of the left subdiaphragmatic space, using silicone tube.
 Conclusion. Organ–preserving operations in splenic injury are expedient to be performed on ІІІ level of the medical help delivery, while presence of local conditions, such as the organ injury degrees І and ІІ,determined in accordance to Moore, as well as adequate time and equipment.