Radiomic analysis of abdominal organs during sepsis of digestive origin in a French intensive care unit.

BACKGROUND: Sepsis is a severe and common cause of admission to the intensive care unit (ICU). Radiomic analysis (RA) may predict organ failure and patient outcomes. The objective of this study was to assess a model of RA and to evaluate its performance in predicting in-ICU mortality and acute kidney injury (AKI) during abdominal sepsis. METHODS: This single-center, retrospective study included patients admitted to the ICU for abdominal sepsis. To predict in-ICU mortality or AKI, elastic net regularized logistic regression and the random forest algorithm were used in a five-fold cross-validation set repeated 10 times. RESULTS: Fifty-five patients were included. In-ICU mortality was 25.5%, and 76.4% of patients developed AKI. To predict in-ICU mortality, elastic net and random forest models, respectively, achieved areas under the curve (AUCs) of 0.48 (95% confidence interval [CI], 0.43-0.54) and 0.51 (95% CI, 0.46-0.57) and were not improved combined with Simplified Acute Physiology Score (SAPS) II. To predict AKI with RA, the AUC was 0.71 (95% CI, 0.66-0.77) for elastic net and 0.69 (95% CI, 0.64-0.74) for random forest, and these were improved combined with SAPS II, respectively; AUC of 0.94 (95% CI, 0.91-0.96) and 0.75 (95% CI, 0.70-0.80) for elastic net and random forest, respectively. CONCLUSIONS: This study suggests that RA has poor predictive performance for in-ICU mortality but good predictive performance for AKI in patients with abdominal sepsis. A secondary validation cohort is needed to confirm these results and the assessed model.

Prediction of cesarean delivery in class III obese nulliparous women: An externally validated model using machine learning.

BACKGROUND: class III obese women, are at a higher risk of cesarean section during labor, and cesarean section is responsible for increased maternal and neonatal morbidity in this population. OBJECTIVE: the objective of this project was to develop a method with which to quantify cesarean section risk before labor. METHODS: this is a multicentric retrospective cohort study conducted on 410 nulliparous class III obese pregnant women who attempted vaginal delivery in two French university hospitals. We developed two predictive algorithms (a logistic regression and a random forest models) and assessed performance levels and compared them. RESULTS: the logistic regression model found that only initial weight and labor induction were significant in the prediction of unplanned cesarean section. The probability forest was able to predict cesarean section probability using only two pre-labor characteristics: initial weight and labor induction. Its performances were higher and were calculated for a cut-point of 49.5% risk and the results were (with 95% confidence intervals): area under the curve 0.70 (0.62,0.78), accuracy 0.66 (0.58, 0.73), specificity 0.87 (0.77, 0.93), and sensitivity 0.44 (0.32, 0.55). CONCLUSIONS: this is an innovative and effective approach to predicting unplanned CS risk in this population and could play a role in the choice of a trial of labor versus planned cesarean section. Further studies are needed, especially a prospective clinical trial. FUNDING: French state funds "Plan Investissements d'Avenir" and Agence Nationale de la Recherche.

Predicting COVID-19 positivity and hospitalization with multi-scale graph neural networks.

The pandemic of COVID-19 is undoubtedly one of the biggest challenges for modern healthcare. In order to analyze the spatio-temporal aspects of the spread of COVID-19, technology has helped us to track, identify and store information regarding positivity and hospitalization, across different levels of municipal entities. In this work, we present a method for predicting the number of positive and hospitalized cases via a novel multi-scale graph neural network, integrating information from fine-scale geographical zones of a few thousand inhabitants. By leveraging population mobility data and other features, the model utilizes message passing to model interaction between areas. Our proposed model manages to outperform baselines and deep learning models, presenting low errors in both prediction tasks. We specifically point out the importance of our contribution in predicting hospitalization since hospitals became critical infrastructure during the pandemic. To the best of our knowledge, this is the first work to exploit high-resolution spatio-temporal data in a multi-scale manner, incorporating additional knowledge, such as vaccination rates and population mobility data. We believe that our method may improve future estimations of positivity and hospitalization, which is crucial for healthcare planning.

Paradigm shift: should preoperative 3D reconstruction models become mandatory before hepatectomy for hepatocellular carcinoma (HCC)? Results of a multicenter prospective trial.

BACKGROUND: A preoperative surgical strategy before hepatectomy for hepatocellular carcinoma is fundamental to minimize postoperative morbidity and mortality and to reach the best oncologic outcomes. Preoperative 3D reconstruction models may help to better choose the type of procedure to perform and possibly change the initially established plan based on conventional 2D imaging. METHODS: A non-randomized multicenter prospective trial with 136 patients presenting with a resectable hepatocellular carcinoma who underwent open or minimally invasive liver resection. Measurement was based on the modification rate analysis between conventional 2D imaging (named "Plan A") and 3D model analysis ("Plan B"), and from Plan B to the final procedure performed (named "Plan C"). RESULTS: The modification rate from Plan B to Plan C (18%) was less frequent than the modification from Plan A to Plan B (35%) (OR = 0.32 [0.15; 0.64]). Concerning secondary objectives, resection margins were underestimated in Plan B as compared to Plan C (-3.10 mm [-5.04; -1.15]). CONCLUSION: Preoperative 3D imaging is associated with a better prediction of the performed surgical procedure for liver resections in HCC, as compared to classical 2D imaging.

MRI for the Detection of Small Bowel Ischemic Injury in Arterial Acute Mesenteric Ischemia: Preclinical Study in a Porcine Model.

BACKGROUND: MRI is the reference for the diagnosis of arterial cerebral ischemia, but its role in acute mesenteric ischemia (AMI) is poorly known. PURPOSE: To assess MRI detection of early ischemic bowel lesions in a porcine model of arterial AMI. STUDY TYPE: Prospective/cohort. ANIMAL MODEL: Porcine model of arterial AMI obtained by embolization of the superior mesenteric artery (seven pigs). FIELD STRENGTH/SEQUENCE: A 5-T. T1 gradient-echo-weighted-imaging (WI), half-Fourier-acquisition-single-shot-turbo-spin-echo, T2 turbo-spin-echo, true-fast-imaging-with-steady-precession (True-FISP), diffusion-weighted-echo-planar (DWI). ASSESSMENT: T1-WI, T2-WI, and DWI were performed before and continuously after embolization for 6 hours. The signal intensity (SI) of the ischemic bowel was assessed visually and quantitatively on all sequences. The apparent diffusion coefficient (ADC) was assessed. STATISTICAL TESTS: Paired Student's t-test or Mann-Whitney U-test, significance at P < 0.05. RESULTS: One pig died from non-AMI-related causes. The remaining pigs underwent a median 5 h53 (range 1 h24-6 h01) of ischemia. Visually, the ischemic bowel showed signal hyperintensity on DWI-b800 after a median 85 (57-276) minutes compared to the nonischemic bowel. DWI-b800 SI significantly increased after 2 hours (+19%) and the ADC significant decrease within the first hour (-31%). The ischemic bowel was hyperintense on precontrast T1-WI after a median 87 (70-171) minutes with no significant quantitative changes over time (P = 0.46-0.93). The ischemic bowel was hyperintense on T2-WI in three pigs with a significant SI increase on True-FISP after 1 and 2 hours. DATA CONCLUSION: Changes in SI and ADC can be seen early after the onset of arterial AMI with DWI. The value of T2-WI appears to be limited. EVIDENCE LEVEL: 1 TECHNICAL EFFICACY: Stage 2.

Preoperative SPECT/CT + intraoperative CT fusion enabling surgical augmented reality to target sentinel lymph node in endometrial cancer.

PURPOSE: To establish a proof-of-concept study using a phantom model to allow the fusion of preoperative single-photon emission computed tomography (SPECT) combined with computed tomography (CT), also known as SPECT/CT, with intraoperative CT, enabling the application of an augmented reality (AR) surgical guidance system for pelvic sentinel lymph node (SLN) detection in endometrial cancer patients. METHODS: A three-dimensional (3D) pelvic phantom model printed in a gelatin-based scaffold including a radiopaque pelvis, a vascular tree mimicking the iliac vessels, two 3D-printed fillable spheres representing the target pelvic sentinel lymph nodes, and a calibration board was developed. A planar with SPECT/CT lymphoscintigraphy and CT were performed independently on the model. We performed all the necessary steps to achieve the fusion between SPECT/CT and CT. Then, we performed a laparoscopy of the pelvic anatomy on the phantom model to assess in real time the overlay of the recording on the anatomical structures and AR guidance system performance. RESULTS: We have successfully completed all the steps needed to fuse the two imaging procedures. This allowed us to apply, in real time, our surgical guidance system with the coverage rate of the visible surface by the augmented reality surface, respectively, on the left SLN 99.48% and on the right SLN 99.42%. CONCLUSION: Co-registration and real-time fusion between a preoperative SPECT/CT and intraoperative CT are feasible. The metric performance of our guidance system is excellent in relation to possible SPECT/CT and CT fusion. Based on our results, we are able to translate the technology to patients, and we initiated a clinical study to evaluate the accuracy of the AR guidance system for endometrial cancer surgery, with a correlation with indocyanine green (ICG)-based technique, representing the gold standard today in the intraoperative detection of SLN in endometrial cancers, despite various limitations.

Analysis of cavitation artifacts in Magnetic Resonance Imaging Thermometry during laser ablation monitoring.

Magnetic Resonance Thermometry Imaging (MRTI) holds great potential in laser ablation (LA) monitoring. It provides the real-time multidimensional visualization of the treatment effect inside the body, thus enabling accurate intraoperative prediction of the thermal damage induced. Despite its great potential., thermal maps obtained with MRTI may be affected by numerous artifacts. Among the sources of error producing artifacts in the images., the cavitation phenomena which could occur in the tissue during LA induces dipole-structured artifacts. In this work., an analysis of the cavitation artifacts occurring during LA in a gelatin phantom in terms of symmetry in space and symmetry of temperature values was performed. Results of 2 Wand 4 W laser power were compared finding higher symmetry for the 2 W case in terms of both dimensions of artifact-lobes and difference in temperature values extracted in specular pixels in the image. This preliminary investigation of artifact features may provide a step forward in the identification of the best strategy to correct and avoid artifact occurrence during thermal therapy monitoring. Clinical Relevance- This work presents an analysis of cavitation artifacts in MRTI from LA which must be corrected to avoid error in the prediction of thermal damage during LA monitoring.

CT features associated with underlying malignancy in patients with diagnosed mesenteric panniculitis.

PURPOSE: The purpose of this study was to identify abdominal computed tomography (CT) features associated with underlying malignancy in patients with mesenteric panniculitis (MP). MATERIALS AND METHODS: This single-institution retrospective longitudinal cohort study included patients with MP and a minimum 1-year abdominopelvic CT follow-up or 2-year clinical follow-up after initial abdominopelvic CT examination. Two radiologists, blinded to patients' medical records, conjointly reviewed CT-based features of MP. Electronic medical records were reviewed for newly diagnosed malignancies with the following specific details: type (lymphoproliferative disease or solid malignancy), location (possible mesenteric drainage or distant), stage, time to diagnosis. An expert panel of three radiologists and one hemato-oncologist, who were blinded to the initial CT-based MP features, assessed the probability of association between MP and malignancy based on the malignancy characteristics. RESULTS: From 2006 to 2016, 444 patients with MP were included. There were 272 men and 172 women, with a median age of 64 years (age range: 25-89); the median overall follow-up was 36 months (IQR: 22, 60; range: 12-170). A total of 34 (8%) patients had a diagnosis of a new malignancy; 5 (1%) were considered possibly related to the MP, all being low-grade B-cell non-Hodgkin lymphomas. CT features associated with the presence of an underlying malignancy were the presence of an MP soft-tissue nodule with a short axis >10 mm (P < 0.0001) or lymphadenopathy in another abdominopelvic region (P < 0.0001). Associating these two features resulted in high diagnostic performance (sensitivity 100%; [95% CI: 57-100]; specificity 99% [95% CI: 98-100]). All related malignancies were identified. CONCLUSION: Further workup to rule out an underlying malignancy is only necessary in the presence of an MP soft-tissue nodule >10 mm or associated abdominopelvic lymphadenopathy.

Artificial Intelligence and Surgery: Ethical Dilemmas and Open Issues.

BACKGROUND: Artificial intelligence (AI) applications aiming to support surgical decision-making processes are generating novel threats to ethical surgical care. To understand and address these threats, we summarize the main ethical issues that may arise from applying AI to surgery, starting from the Ethics Guidelines for Trustworthy Artificial Intelligence framework recently promoted by the European Commission. STUDY DESIGN: A modified Delphi process has been employed to achieve expert consensus. RESULTS: The main ethical issues that arise from applying AI to surgery, described in detail here, relate to human agency, accountability for errors, technical robustness, privacy and data governance, transparency, diversity, non-discrimination, and fairness. It may be possible to address many of these ethical issues by expanding the breadth of surgical AI research to focus on implementation science. The potential for AI to disrupt surgical practice suggests that formal digital health education is becoming increasingly important for surgeons and surgical trainees. CONCLUSIONS: A multidisciplinary focus on implementation science and digital health education is desirable to balance opportunities offered by emerging AI technologies and respect for the ethical principles of a patient-centric philosophy.

Robotically assisted augmented reality system for identification of targeted lymph nodes in laparoscopic gynecological surgery: a first step toward the identification of sentinel node : Augmented reality in gynecological surgery.

BACKGROUND: To prove feasibility of multimodal and temporal fusion of laparoscopic images with preoperative computed tomography scans for a real-time in vivo-targeted lymph node (TLN) detection during minimally invasive pelvic lymphadenectomy and to validate and enable such guidance for safe and accurate sentinel lymph node dissection, including anatomical landmarks in an experimental model. METHODS: A measurement campaign determined the most accurate tracking system (UR5-Cobot versus NDI Polaris). The subsequent interventions on two pigs consisted of an identification of artificial TLN and anatomical landmarks without and with augmented reality (AR) assistance. The AR overlay on target structures was quantitatively evaluated. The clinical relevance of our system was assessed via a questionnaire completed by experienced and trainee surgeons. RESULTS: An AR-based robotic assistance system that performed real-time multimodal and temporal fusion of laparoscopic images with preoperative medical images was developed and tested. It enabled the detection of TLN and their surrounding anatomical structures during pelvic lymphadenectomy. Accuracy of the CT overlay was > 90%, with overflow rates < 6%. When comparing AR to direct vision, we found that scores were significatively higher in AR for all target structures. AR aided both experienced surgeons and trainees, whether it was for TLN, ureter, or vessel identification. CONCLUSION: This computer-assisted system was reliable, safe, and accurate, and the present achievements represent a first step toward a clinical study.

Improved Detection of Chronic Obstructive Pulmonary Disease at Chest CT Using the Mean Curvature of Isophotes.

PURPOSE: To determine if the mean curvature of isophotes (MCI), a standard computer vision technique, can be used to improve detection of chronic obstructive pulmonary disease (COPD) at chest CT. MATERIALS AND METHODS: In this retrospective study, chest CT scans were obtained in 243 patients with COPD and 31 controls (among all 274: 151 women [mean age, 70 years; range, 44-90 years] and 123 men [mean age, 71 years; range, 29-90 years]) from two community practices between 2006 and 2019. A convolutional neural network (CNN) architecture was trained on either CT images or CT images transformed through the MCI algorithm. Separately, a linear classification based on a single feature derived from the MCI computation (called hMCI1) was also evaluated. All three models were evaluated with cross-validation, using precision-macro and recall-macro metrics, that is, the mean of per-class precision and recall values, respectively (the latter being equivalent to balanced accuracy). RESULTS: Linear classification based on hMCI1 resulted in a higher recall-macro relative to the CNN trained and applied on CT images (0.85 [95% CI: 0.84, 0.86] vs 0.77 [95% CI: 0.75, 0.79]) but with a similar reduction in precision-macro (0.66 [95% CI: 0.65, 0.67] vs 0.77 [95% CI: 0.75, 0.79]). The CNN model trained and applied on MCI-transformed images had a higher recall-macro (0.85 [95% CI: 0.83, 0.87] vs 0.77 [95% CI: 0.75, 0.79]) and precision-macro (0.85 [95% CI: 0.83, 0.87] vs 0.77 [95% CI: 0.75, 0.79]) relative to the CNN trained and applied on CT images. CONCLUSION: The MCI algorithm may be valuable toward the automated detection and diagnosis of COPD on chest CT scans as part of a CNN-based pipeline or with stand-alone features.Keywords: Chronic Obstructive Pulmonary Disease, Quantification, Lung, CT Supplemental material is available for this article. See also the invited commentary by Vannier in this issue.© RSNA, 2021.

Conventional and artificial intelligence-based imaging for biomarker discovery in chronic liver disease.

Chronic liver diseases, resulting from chronic injuries of various causes, lead to cirrhosis with life-threatening complications including liver failure, portal hypertension, hepatocellular carcinoma. A key unmet medical need is robust non-invasive biomarkers to predict patient outcome, stratify patients for risk of disease progression and monitor response to emerging therapies. Quantitative imaging biomarkers have already been developed, for instance, liver elastography for staging fibrosis or proton density fat fraction on magnetic resonance imaging for liver steatosis. Yet, major improvements, in the field of image acquisition and analysis, are still required to be able to accurately characterize the liver parenchyma, monitor its changes and predict any pejorative evolution across disease progression. Artificial intelligence has the potential to augment the exploitation of massive multi-parametric data to extract valuable information and achieve precision medicine. Machine learning algorithms have been developed to assess non-invasively certain histological characteristics of chronic liver diseases, including fibrosis and steatosis. Although still at an early stage of development, artificial intelligence-based imaging biomarkers provide novel opportunities to predict the risk of progression from early-stage chronic liver diseases toward cirrhosis-related complications, with the ultimate perspective of precision medicine. This review provides an overview of emerging quantitative imaging techniques and the application of artificial intelligence for biomarker discovery in chronic liver disease.

Artificial intelligence-based radiomics models in endometrial cancer: A systematic review.

BACKGROUND: Radiological preoperative assessment of endometrial cancer (EC) is in some cases not precise enough and its performances improvement could lead to a clinical benefit. Radiomics is a recent field of application of artificial intelligence (AI) in radiology. AIMS: To investigate the contribution of radiomics on the radiological preoperative assessment of patients with EC; and to establish a simple and reproducible AI Quality Score applicable to Machine Learning and Deep Learning studies. METHODS: We conducted a systematic review of current literature including original articles that studied EC through imaging-based AI techniques. Then, we developed a novel Simplified and Reproducible AI Quality score (SRQS) based on 10 items which ranged to 0 to 20 points in total which focused on clinical relevance, data collection, model design and statistical analysis. SRQS cut-off was defined at 10/20. RESULTS: We included 17 articles which studied different radiological parameters such as deep myometrial invasion, lympho-vascular space invasion, lymph nodes involvement, etc. One article was prospective, and the others were retrospective. The predominant technique was magnetic resonance imaging. Two studies developed Deep Learning models, while the others machine learning ones. We evaluated each article with SRQS by 2 independent readers. Finally, we kept only 7 high-quality articles with clinical impact. SRQS was highly reproducible (Kappa = 0.95 IC 95% [0.907-0.988]). CONCLUSION: There is currently insufficient evidence on the benefit of radiomics in EC. Nevertheless, this field is promising for future clinical practice. Quality should be a priority when developing these new technologies.

COVID-19 Efforts at the Institute of Image Guided Surgery (IHU-Strasbourg): 2020.

We submit a summary of some of the activities of the IHU-Strasbourg during the initial period of the COVID-19 pandemic. These were presented as part of the coronnavation effort coordinated by Dr Adrian Park. Three initiatives are presented as follows: Protect-Est App, healthcare worker stress, and converted diving mask for ventilation. Two of the 3 projects are still ongoing, and one (Predoict-Est) has been adopted nationally.

Diffuse Versus Localized Caroli Disease: A Comparative MRCP Study.

OBJECTIVE. The purpose of this multicenter retrospective study was to assess the MRCP features of Caroli disease (CD). MATERIALS AND METHODS. Sixty-six patients were identified from 2000 to 2019. The inclusion criteria were diagnosis of diffuse or localized CD mentioned in an imaging report, presence of intrahepatic bile duct (IHBD) dilatation, and having undergone an MRCP examination. The exclusion criteria included presence of obstructive proximal biliary stricture and having undergone hepatobiliary surgery other than cholecystectomy. Histopathology records were available for 53 of the 66 (80%) patients. Diffuse and localized diseases were compared by chi-square and t tests and Kaplan-Meier model. RESULTS. Forty-five patients had diffuse bilobar CD ((five pediatric patients [three girls and two boys] with a mean [± SD] age of 8 ± 5 years [range, 1-15 years] and 40 adult patients [26 men and 14 women] with a mean age of 35 ± 11 years [range, 20-62 years]) and 21 patients had localized disease (12 men and 9 women; mean age, 54 ± 14 years). Congenital hepatic fibrosis was found only in patients with diffuse CD (35/45 [78%]), as was a "central dot" sign (15/35 [43%]). IHBD dilatation with both saccular and fusiform features was found in 43 (96%) and the peripheral "funnel-shaped" sign in 41 (91%) of the 45 patients with diffuse CD but in none of the patients with localized disease (p < .001). Intrahepatic biliary calculi were found in all patients with localized disease but in only 16 of the 45 (36%) patients with diffuse CD (p < .001). Left liver atrophy was found in 18 of the 21 (86%) patients with localized disease and in none of the patients with diffuse CD (p < .001). The overall survival rate among patients with diffuse CD was significantly lower than that among patients with localized disease (p = .03). CONCLUSION. Diffuse IHBD dilatation with both saccular and fusiform features associated with the peripheral funnel-shaped sign can be used for the diagnosis of CD on MRCP. Localized IHBD dilatation seems to be mainly related to primary intrahepatic lithiasis.

Precision image-guided colonic surgery: proof of concept for enhanced preoperative and intraoperative vascular imaging.

BACKGROUND: Colorectal surgery has benefited from advances in precision medicine such as total mesorectal resection, and recently, mesocolon resection, fluorescent perfusion imaging, and fluorescent node mapping. However, these advances fail to address the variable quality of mesocolon dissection and the directed extent of vascular dissection (including high ligation) or pre-resection anastomotic perfusion mapping, thereby impacting anastomotic leaks. We propose a new paradigm of precision image-directed colorectal surgery involving 3D preoperative resection modeling and intraoperative fluoroscopic and fluorescence vascular imaging which better defines optimal dissection planes and vascular vs. anatomy-based resection lines according to our hypothesis. METHODS: Six pigs had preoperative CT with vascular 3D reconstruction allowing for the preoperative planning of vascular-based dissection. Laparoscopic surgery was performed in a hybrid operating room (OR). Superselective arterial catheterization was performed in branches of the superior mesenteric artery (SMA) or the inferior mesenteric artery (IMA). Intraoperative boluses of 0.1 mg/kg or a continuous infusion of indocyanine green (ICG) (0.01 mg/mL) were administered to guide fluorescent-based sigmoid and ileocecal resections. Fluorescence was assessed using proprietary software at several regions of interest (ROI) in the right and left colon. RESULTS: The approach was feasible and safe. Selective catheterization took an average of 43 min. Both bolus and continuous perfusion clearly marked pre-identified vessels (arteries/veins) and the target colon segment, facilitating precise resections based on the visible vascular anatomy. Quantitative software analysis indicated the optimal resection margin for each ROI. CONCLUSION: Intra-arterial fluorescent mapping allows visualization of major vascular structures and segmental colonic perfusion. This may help to prevent any inadvertent injury to major vascular structures and to precisely determine perfusion-based resection planes and margins. This could enable tailoring of the amount of colon resected, ensure good anastomotic perfusion, and improve oncological outcomes.

Diagnostic Accuracy and Clinical Impact of Sentinel Lymph Node Sampling in Endometrial Cancer at High Risk of Recurrence: A Meta-Analysis.

PURPOSE: To assess the value of sentinel lymph node (SLN) sampling in high risk endometrial cancer according to the ESMO-ESGO-ESTRO classification. METHODS: We performed a comprehensive search on PubMed for clinical trials evaluating SLN sampling in patients with high risk endometrial cancer: stage I endometrioid, grade 3, with at least 50% myometrial invasion, regardless of lymphovascular space invasion status; or stage II; or node-negative stage III endometrioid, no residual disease; or non-endometrioid (serous or clear cell or undifferentiated carcinoma, or carcinosarcoma). All patients underwent SLN sampling followed by pelvic with or without para-aortic lymphadenectomy. RESULTS: We included 17 original studies concerning 1322 women. Mean detection rates were 89% for unilateral and 68% for bilateral. Pooled sensitivity was 88.5% (95%CI: 81.2-93.2%), negative predictive value was 96.0% (95%CI: 93.1-97.7%), and false negative rate was 11.5% (95%CI: 6.8; 18.8%). We noted heterogeneity in SLN techniques between studies, concerning the tracer and its detection, the injection site, the number of injections, and the surgical approach. Finally, we found a correlation between the number of patients included and the SLN sampling performances. DISCUSSION: This meta-analysis estimated the SLN sampling performances in high risk endometrial cancer patients. Data from the literature show the feasibility, the safety, the limits, and the impact on surgical de-escalation of this technique. In conclusion, our study supports the hypothesis that SLN sampling could be a valuable technique to diagnose lymph node involvement for patients with high risk endometrial cancer in replacement of conventional lymphadenectomy. Consequently, randomized clinical trials are necessary to confirm this hypothesis.

Multimodality Imaging and Artificial Intelligence for Tumor Characterization: Current Status and Future Perspective.

Research in medical imaging has yet to do to achieve precision oncology. Over the past 30 years, only the simplest imaging biomarkers (RECIST, SUV,…) have become widespread clinical tools. This may be due to our inability to accurately characterize tumors and monitor intratumoral changes in imaging. Artificial intelligence, through machine learning and deep learning, opens a new path in medical research because it can bring together a large amount of heterogeneous data into the same analysis to reach a single outcome. Supervised or unsupervised learning may lead to new paradigms by identifying unrevealed structural patterns across data. Deep learning will provide human-free, undefined upstream, reproducible, and automated quantitative imaging biomarkers. Since tumor phenotype is driven by its genotype and thus indirectly defines tumoral progression, tumor characterization using machine learning and deep learning algorithms will allow us to monitor molecular expression noninvasively, anticipate therapeutic failure, and lead therapeutic management. To follow this path, quality standards have to be set: standardization of imaging acquisition as it has been done in the field of biology, transparency of the model development as it should be reproducible by different institutions, validation, and testing through a high-quality process using large and complex open databases and better interpretability of these algorithms.