Artificial Intelligence and Computer Vision during Surgery: Discussing Laparoscopic Images with ChatGPT4—Preliminary Results

Introduction: Ultrafast latest developments in artificial intelligence (ΑΙ) have recently multiplied concerns regarding the future of robotic autonomy in surgery. However, the literature on the topic is still scarce. Aim: To test a novel AI commercially available tool for image analysis on a series of laparoscopic scenes. Methods: The research tools included OPENAI CHATGPT 4.0 with its corresponding image recognition plugin which was fed with a list of 100 laparoscopic selected snapshots from common surgical procedures. In order to score reliability of received responses from image-recognition bot, two corresponding scales were developed ranging from 0 - 5. The set of images was divided into two groups: unlabeled (Group A) and labeled (Group B), and according to the type of surgical procedure or image resolution. Results: AI was able to recognize correctly the context of surgical-related images in 97% of its reports. For the labeled surgical pictures, the image-processing bot scored 3.95/5 (79%), whilst for the unlabeled, it scored 2.905/5 (58.1%). Phases of the procedure were commented in detail, after all successful interpretations. With rates 4 - 5/5, the chatbot was able to talk in detail about the indications, contraindications, stages, instrumentation, complications and outcome rates of the operation discussed. Conclusion: Interaction between surgeon and chatbot appears to be an interesting frontend for further research by clinicians in parallel with evolution of its complex underlying infrastructure. In this early phase of using artificial intelligence for image recognition in surgery, no safe conclusions can be drawn by small cohorts with commercially available software. Further development of medically-oriented AI software and clinical world awareness are expected to bring fruitful information on the topic in the years to come.

Lung imaging characteristics in a patient infected with Elizabethkingia miricola following cerebral hemorrhage surgery: A case report

BACKGROUND Elizabethkingia miricola is a non-fermenting gram-negative bacterium,which was first isolated from the condensate of the Russian peace space station in 2003.Most studies on this bacterium have been carried out in the laboratory,and clinical case studies are rare.To date,a total of 6 clinical cases have been reported worldwide.CASE SUMMARY We present the first case of postoperative pulmonary infection in a patient with intracerebral hemorrhage due to Elizabethkingia miricola.The imaging character-istics of pulmonary infection were identified and the formulation and selection of the clinical treatment plan for this patient are discussed.CONCLUSION Elizabethkingia miricola infection is rare.When pulmonary infection occurs,computed tomography imaging may show diffuse distribution of a ground glass density shadow in both lungs,the air containing bronchial sign in local areas,thickening of bronchial vascular bundle,and pleural effusion.

Relational graph location network for multi-view image localization

In multi-view image localization task,the features of the images captured from different views should be fused properly.This paper considers the classification-based image localization problem.We propose the relational graph location network(RGLN)to perform this task.In this network,we propose a heterogeneous graph construction approach for graph classification tasks,which aims to describe the location in a more appropriate way,thereby improving the expression ability of the location representation module.Experiments show that the expression ability of the proposed graph construction approach outperforms the compared methods by a large margin.In addition,the proposed localization method outperforms the compared localization methods by around 1.7%in terms of meter-level accuracy.

Imaged guided surgery during arteriovenous malformation of gastrointestinal stromal tumor using hyperspectral and indocyanine green visualization techniques:A case report

BACKGROUND This case report demonstrates the simultaneous development of a gastrointestinal stromal tumour(GIST)with arteriovenous malformations(AVMs)within the jejunal mesentery.A 74-year-old male presented to the department of surgery at our institution with a one-month history of abdominal pain.Contrast-enhanced computed tomography revealed an AVM.During exploratory laparotomy,hyperspectral imaging(HSI)and indocyanine green(ICG)fluorescence were used to evaluate the extent of the tumour and determine the resection margins.Intraoperative imaging confirmed AVM,while histopathological evaluation showed an epithelioid,partially spindle cell GIST.CASE SUMMARY This is the first case reporting the use of HSI and ICG to image GIST intermingled with an AVM.The resection margins were planned using intraoperative analysis of additional optical data.Image-guided surgery enhances the clinician’s knowledge of tissue composition and facilitates tissue differentiation.CONCLUSION Since image-guided surgery is safe,this procedure should increase in popularity among the next generation of surgeons as it is associated with better postoperative outcomes.

Recent advances in computerized imaging and its vital roles in liverdisease diagnosis, preoperative planning, and interventional liversurgery: A review

The earliest and most accurate detection of the pathological manifestations of hepatic diseases ensures effective treatments and thus positive prognostic outcomes.In clinical settings,screening and determining the extent of a pathology are prominent factors in preparing remedial agents and administering approp-riate therapeutic procedures.Moreover,in a patient undergoing liver resection,a realistic preoperative simulation of the subject-specific anatomy and physiology also plays a vital part in conducting initial assessments,making surgical decisions during the procedure,and anticipating postoperative results.Conventionally,various medical imaging modalities,e.g.,computed tomography,magnetic resonance imaging,and positron emission tomography,have been employed to assist in these tasks.In fact,several standardized procedures,such as lesion detection and liver segmentation,are also incorporated into prominent commercial software packages.Thus far,most integrated software as a medical device typically involves tedious interactions from the physician,such as manual delineation and empirical adjustments,as per a given patient.With the rapid progress in digital health approaches,especially medical image analysis,a wide range of computer algorithms have been proposed to facilitate those procedures.They include pattern recognition of a liver,its periphery,and lesion,as well as pre-and postoperative simulations.Prior to clinical adoption,however,software must conform to regulatory requirements set by the governing agency,for instance,valid clinical association and analytical and clinical validation.Therefore,this paper provides a detailed account and discussion of the state-of-the-art methods for liver image analyses,visualization,and simulation in the literature.Emphasis is placed upon their concepts,algorithmic classifications,merits,limitations,clinical considerations,and future research trends.

Advances in the application of auxiliary imaging techniques in parathyroid diseases

Hypoparathyroidism is one of the main complications after total thyroidectomy,severely affecting patients’quality of life.How to effectively protect parathyroid function after surgery and reduce the incidence of hypoparathyroidism has always been a key research area in thyroid surgery.Therefore,precise localization of parathyroid glands during surgery,effective imaging,and accurate surgical resection have become hot topics of concern for thyroid surgeons.In response to this clinical phenomenon,this study compared several different imaging methods for parathyroid surgery,including nanocarbon,indocyanine green,near-infrared imaging techniques,and technetium-99m methoxyisobutylisonitrile combined with gamma probe imaging technology.The advantages and disadvantages of each method were analyzed,providing scientific recommendations for future parathyroid imaging.In recent years,some related basic and clinical research has also been conducted in thyroid surgery.This article reviewed relevant literature and provided an overview of the practical application progress of various imaging techniques in parathyroid surgery.

Application Research of PETD Combined with MRI Nerve Root Water Imaging in the Minimally Invasive Treatment of LDH

Objective: This study aims to evaluate the safety and efficacy of PETD combined with nerve root water imaging of MRI for the treatment of lumbar disc herniation. Methods: A retrospective review was performed on 62 patients with lumbar disc herniation from March 2019 to March 2021. The study included an experimental group of 30 patients and a control group of 32 patients. The experimental group underwent PETD combined with nerve root water imaging of MRI, while the control group received traditional PETD treatment. The visual analogue scoring method (VAS score), and JOA lumbar spine function score before and after surgery were compared between the two groups, and efficacy was assessed and compared using the MacNab score. Results: The mean operation time was significantly reduced in the experimental group (56.43 ±10.40 minutes) compared to the control group (65.69 ±14.12 minutes). The VAS score was compared between the two groups with preoperative (p = 0.624), one month after surgery (p = 0.325), three months after surgery (p = 0.676), one year after surgery (p = 0.341);The JOA score was compared between the two groups with preoperative (p = 0.961), one month after the surgery (p = 0.266), three months after surgery (p = 0.185), one year after surgery (p = 0.870), they were no significant statistical difference;The efficacy evaluation of the last follow-up Macnab showed that all the 30 patients in the experimental group were excellent, 31 of 32 patients in the control group were excellent, 1 case was good;There was no statistical difference in the comparison between the two groups (p > 0.05). Conclusion: The study concludes that the combined approach of PETD with nerve root water imaging of MRI is a safe, effective, and more efficient alternative to conventional PETD for treating lumbar disc herniation.

Magnetic resonance imaging findings of radiation-induced breast angiosarcoma:A case report

BACKGROUND Breast conservation surgery(BCS)with adjuvant radiotherapy has become a gold standard in the treatment of early-stage breast cancer,significantly reducing the risk of tumor recurrence.However,this treatment is associated with adverse effects,including the rare but aggressive radiation-induced angiosarcoma(RIAS).Despite its rarity and nonspecific initial presentation,RIAS presents a challenging diagnosis,emphasizing the importance of imaging techniques for early detection and accurate diagnosis.CASE SUMMARY We present a case of a 48-year-old post-menopausal woman who developed skin ecchymosis on the right breast seven years after receiving BCS and adjuvant radiotherapy for breast cancer.Initial mammography and ultrasound were inconclusive,showing post-treatment changes but failing to identify the underlying angiosarcoma.Contrast-enhanced breast magnetic resonance imaging(MRI)revealed diffuse skin thickening and nodularity with distinctive enhan-cement kinetics,leading to the diagnosis of RIAS.This case highlights the crucial role of MRI in diagnosing and determining the extent of RIAS,facilitating timely and appropriate surgical intervention.CONCLUSION Breast MRI is crucial for detecting RIAS,especially when mammography and ultrasound are inconclusive.

indocyanine green-based fluorescence imaging in visceral and hepatobiliary and pancreatic surgery:State of the art and future directions

In recent years, the use of fluorescence-guided surgery(FGS) to treat benign and malignant visceral, hepatobiliary and pancreatic neoplasms has significantly increased. FGS relies on the fluorescence signal emitted by injected substances(fluorophores) after being illuminated by ad hoc laser sources to help guide the surgical procedure and provide the surgeon with real-time visualization of the fluorescent structures of interest that would be otherwise invisible. This review surveys and discusses the most common and emerging clinical applications of indocyanine green(ICG)-based fluorescence in visceral, hepatobiliary and pancreatic surgery. The analysis, findings, and discussion presented here rely on the authors' significant experience with this technique in their medical institutions, an up-to-date review of the most relevant articles published on this topic between 2014 and 2018, and lengthy discussions with key opinion leaders in the field during recent conferences and congresses. For each application, the benefits and limitations of this technique, as well as applicable future directions, are described. The imaging of fluorescence emitted by ICG is a simple, fast,relatively inexpensive, and harmless tool with numerous different applications in surgery for both neoplasms and benign pathologies of the visceral and hepatobiliary systems. The ever-increasing availability of visual systems that can utilize this tool will transform some of these applications into the standard of care in the near future. Further studies are needed to evaluate the strengths and weaknesses of each application of ICG-based fluorescence imaging in abdominal surgery.

3D Surface Reconstruction of Coarse Aggregate Particles from Occlusion-Free Multi-View Images

Rapidly and accurately assessing the geometric characteristics of coarse aggregate particles is crucial for ensuring pavement performance in highway engineering.This article introduces an innovative system for the three-dimensional(3D)surface reconstruction of coarse aggregate particles using occlusion-free multi-view imaging.The system captures synchronized images of particles in free fall,employing a matte sphere and a nonlinear optimization approach to estimate the camera projection matrices.A pre-trained segmentation model is utilized to eliminate the background of the images.The Shape from Silhouettes(SfS)algorithm is then applied to generate 3D voxel data,followed by the Marching Cubes algorithm to construct the 3D surface contour.Validation against standard parts and diverse coarse aggregate particles confirms the method's high accuracy,with an average measurement precision of 0.434 mm and a significant increase in scanning and reconstruction efficiency.

Added value of indocyanine green fluorescence imaging in liver surgery

Recently,indocyanine green(ICG)fluorescence imaging has been widely used as a substitute for cholangiography in hepatobiliary surgery,to detect hepatic tumors,for accurate anatomical hepatectomy,and to increase the safety and accuracy of minimally invasive(laparoscopic and robotic)hepatectomy.The clinical relevance of this method has been increasing gradually,as new procedures develop in this field.Various important roles and the latest added value of ICG fluorescence imaging in liver surgery are discussed in this report.

Application value analysis of magnetic resonance imaging and computed tomography in the diagnosis of intracranial infection after craniocerebral surgery

BACKGROUND Intracranial infection is a common clinical disease.Computed tomography(CT)and magnetic resonance imaging(MRI)have certain sensitivity and have good diagnostic efficacy.AIM To study the application value of MRI and CT in the diagnosis of intracranial infection after craniocerebral surgery.METHODS We selected 82 patients who underwent craniocerebral surgery(including 40 patients with intracranial infection and 42 patients without infection)during the period from April 2016 to June 2019 in our hospital.All 82 patients received CT and MRI examinations,and their clinical data were reviewed.A retrospective analysis was performed,and the coincidence rate of positive diagnosis and the overall diagnosis coincidence rate of different pathogenic infection types were measured with the two examination methods.The diagnostic sensitivity and specificity as well as the positive and negative predictive values of the two examination methods were compared.RESULTS For all types of pathogenic infections(Staphylococcus aureus,Staphylococcus hemolyticus,Staphylococcus epidermidis,and others),MRI scans had higher positive diagnostic coincidence rates than CT scans;the overall diagnostic coincidence rate,sensitivity,specificity,positive predictive value,and negative predictive values were significantly higher with MRI examinations than with CT examinations,and the differences were statistically significant(P<0.05).CONCLUSION MRI examination can accurately diagnose intracranial infection after clinical craniocerebral surgery.Compared with CT,MRI had higher diagnostic efficiency.The diagnostic sensitivity and specificity,the diagnostic coincidence rate,and the positive and negative predictive values were significantly higher with MRI than with conventional CT,which can be actively promoted.

Three-dimensional,isotropic imaging of mouse brain using multi-view deconvolution light sheet microscopy

We present a threedimensional(3D)isotropic imaging of mouse brain using light-sheet fuo-rescent microscopy(LSFM)in conjumction with a multi-view imaging computation.Unlike common single view LSFM is used for mouse brain imaging,the brain tissue is 3D imaged under eight views in our study,by a home-built selective plane ilumination microscopy(SPIM).An output image containing complete structural infornation as well as significantly improved res olution(~4 times)are then computed based on these eight views of data,using a bead-guided multi-view registration and deconvolution.With superior imaging quality,the astrocyte and pyrarmidal neurons together with their subcellular nerve fbers can be clearly visualized and segmented.With further incuding other computational methods,this study can be potentially scaled up to map the conectome of whole mouse brain with a simple light.sheet microscope.

Imaging of the treated breast post breast conservation surgery/oncoplasty: Pictorial review

Mammographic appearance of the normal breast is altered in the post-operative setting. It is essential to be aware of the normal findings as well as to identify features of recurrent disease with particular emphasis on radiologicalpathological concordance. Digital breast tomosynthesis and volumetric breast density add incremental value in this clinical setting. We present a pictorial review of various cases to illustrate normal post-operative findings as well as mammographic features suspicious for recurrent disease.

Image-guided lens extraction surgery: a systematic review

A systematic review of the recent literature regarding the current image-guided systems used for cataract surgery or refractive lens exchange was performed based on the PubMed and Google Scholar databases in March 2018. Literature review returned 21 eligible studies. These studies compared image-guided systems with other keratometric devices regarding their accuracy, repeatability and reproducibility in measurement of keratometric values, astigmatism magnitude and axis, as well as in IOL power calculation. Additionally, the image-guided systems were compared with conventional manual ink-marking techniques for the alignment of toric IOLs. In conclusion, image-guided systems seem to be an accurate and reliable technology with measurements of high repeatability and reproducibility regarding the keratometry and IOL power calculation, but not yet interchangeable with the current established and validated keratometric devices. However, they are superior over the conventional manual inkmarking techniques for toric IOL alignment.

Microstructural morphology and visual acuity outcome in eyes with epiretinal membrane before,during,and after membrane peeling in intraoperative optical coherence tomography assisted macular surgery

AIM:To measure the difference of intraoperative central macular thickness(CMT)before,during,and after membrane peeling and investigate the influence of intraoperative macular stretching on postoperative best corrected visual acuity(BCVA)outcome and postoperative CMT development.METHODS:A total of 59 eyes of 59 patients who underwent vitreoretinal surgery for epiretinal membrane was analyzed.Videos with intraoperative optical coherence tomography(OCT)were recorded.Difference of intraoperative CMT before,during,and after peeling was measured.Pre-and postoperatively obtained BCVA and spectral-domain OCT images were analyzed.RESULTS:Mean age of the patients was 70±8.13y(range 46-86y).Mean baseline BCVA was 0.49±0.27 log MAR(range 0.1-1.3).Three and six months postoperatively the mean BCVA was 0.36±0.25(P=0.01 vs baseline)and 0.38±0.35(P=0.08 vs baseline)log MAR respectively.Mean stretch of the macula during surgery was 29%from baseline(range 2%-159%).Intraoperative findings of macular stretching did not correlate with visual acuity outcome within 6mo after surgery(r=-0.06,P=0.72).However,extent of macular stretching during surgery significantly correlated with less reduction of CMT at the fovea centralis(r=-0.43,P<0.01)and 1 mm nasal and temporal from the fovea(r=-0.37,P=0.02 and r=-0.50,P<0.01 respectively)3mo postoperatively.CONCLUSION:The extent of retinal stretching during membrane peeling may predict the development of postoperative central retinal thickness,though there is no correlation with visual acuity development within the first 6mo postoperatively.

Do Neuroimaging Results Impact Prognosis of Epilepsy Surgery? A Meta-analysis

The neuroimaging results of drug-resistant epilepsy patients play an important role in the surgery decision and prognosis. The aim of this study was to evaluate the impact of these results on the efficacy of epilepay surgery, and then to explore surgical benefit for epilepsy patients with negative magnetic resonance （MR） images. Twenty-four subgroups describing the outcomes of 1475 epilepsy pa- tients with positive-neuroimaging results and 696 patients with negative-neuroimaging results were in- volved in the meta-analysis. Overall, the odds of postoperational seizurefree rate were 2.03 times higher in magnetic resonance imaging-positive （MRI-positive） patients than in MRI-negative patients [odds ratio （OR）=2.03, 95% CI （1.67, 2.47）, P〈0.00001]. For patients with temporal lobe epilepsy （TLE）, the odds were 1.76 times higher in those with MRI-positive results than in those with MRI-negative results [OR=1.76, 95% CI （1.34, 2.32）, P〈0.0001]. For patients with extra-temporal lobe epilepsy （extra-TLE）, the odds were 2.88 times higher in MRI-positive patients than in MRI-negative patients [OR=2.88, 95% CI （1.53, 5.43）, P=0.001]. It was concluded that the seizure-free rate of MRI-positive patients after surgery was higher than that of MRI-negative patients. For patients with negative results, an appropriate surgery should be concerned for TLE.

MARKER-BASED MAPPING BETWEEN MEDICAL IMAGE AND ROBOT SPACE IN ROBOT-ASSISTED SURGERY

This paper focuses on the problems of matching a virtual and a real environments by means of hardware and software tools. The real space is represented by a patient’s bone where a set of cuts by means of robot system is to be made. The virtual space is a 3D model of the bone reconstructed from a set of CT slices. Robot system is then not only to machine bones but also to perform the fundamental step of registration between the two spaces. An external force sensor is used to adjust robot stiffness in order to perform the tactile searching necessary for the registration. A simple but reliable software algorithm is used to control the robot for matching between medical image and robot space in robot-assisted surgery. The results show the system proposed is precise enough for application, and tests been made also clarify the way to improve it.

Magnetic resonance imaging-based deep learning model to predict multiple firings in double-stapled colorectal anastomosis

BACKGROUND Multiple linear stapler firings during double stapling technique(DST)after laparoscopic low anterior resection(LAR)are associated with an increased risk of anastomotic leakage(AL).However,it is difficult to predict preoperatively the need for multiple linear stapler cartridges during DST anastomosis.AIM To develop a deep learning model to predict multiple firings during DST anastomosis based on pelvic magnetic resonance imaging(MRI).METHODS We collected 9476 MR images from 328 mid-low rectal cancer patients undergoing LAR with DST anastomosis,which were randomly divided into a training set(n=260)and testing set(n=68).Binary logistic regression was adopted to create a clinical model using six factors.The sequence of fast spin-echo T2-weighted MRI of the entire pelvis was segmented and analyzed.Pure-image and clinical-image integrated deep learning models were constructed using the mask region-based convolutional neural network segmentation tool and three-dimensional convolutional networks.Sensitivity,specificity,accuracy,positive predictive value(PPV),and area under the receiver operating characteristic curve(AUC)was calculated for each model.RESULTS The prevalence of≥3 linear stapler cartridges was 17.7%(58/328).The prevalence of AL was statistically significantly higher in patients with≥3 cartridges compared to those with≤2 cartridges(25.0%vs 11.8%,P=0.018).Preoperative carcinoembryonic antigen level>5 ng/mL(OR=2.11,95%CI 1.08-4.12,P=0.028)and tumor size≥5 cm(OR=3.57,95%CI 1.61-7.89,P=0.002)were recognized as independent risk factors for use of≥3 linear stapler cartridges.Diagnostic performance was better with the integrated model(accuracy=94.1%,PPV=87.5%,and AUC=0.88)compared with the clinical model(accuracy=86.7%,PPV=38.9%,and AUC=0.72)and the image model(accuracy=91.2%,PPV=83.3%,and AUC=0.81).CONCLUSION MRI-based deep learning model can predict the use of≥3 linear stapler cartridges during DST anastomosis in laparoscopic LAR surgery.This model might help determine the best anastomosis strategy by avoiding DST when there is a high probability of the need for≥3 linear stapler cartridges.

Indocyanine green dye and its application in gastrointestinal surgery:The future is bright green

Indocyanine green(ICG)is a water-soluble fluorescent dye that is minimally toxic and widely used in gastrointestinal surgery.ICG facilitates anatomical identi-fication of structures(e.g.,ureters),assessment of lymph nodes,biliary mapping,organ perfusion and anastomosis assessment,and aids in determining the adequacy of oncological margins.In addition,ICG can be conjugated to artificially created antibodies for tumour markers,such as carcinoembryonic antigen for colorectal,breast,lung,and gastric cancer,prostate-specific antigen for prostate cancer,and cancer antigen 125 for ovarian cancer.Although ICG has shown promising results,the optimization of patient factors,dye factors,equipment,and the method of assessing fluorescence intensity could further enhance its utility.This review summarizes the clinical application of ICG in gastrointestinal surgery and discusses the emergence of novel dyes such as ZW-800 and VM678 that have demonstrated appropriate pharmacokinetic properties and improved target-to-background ratios in animal studies.With the emergence of robotic technology and the increasing reporting of ICG utility,a comprehensive review of clinical application of ICG in gastrointestinal surgery is timely and this review serves that aim.