Rehabilitation is recognised as a cornerstone of intensive care, essential for optimising functional recovery and reducing long-term disability. Contemporary ICU populations, characterised by advanced age, multimorbidity, and prolonged stays, are at heightened risk of muscle wasting, immobility, frailty, cognitive decline, and functional dependence. Mitigation of these sequelae requires careful interprofessional collaboration for person-centred rehabilitation across the care continuum. This review synthesises evidence from randomised controlled trials, meta-analyses, and clinical practice guidelines on rehabilitation during and after intensive care. Best practice within the ICU begins with early awakening and mobilisation with evidence demonstrating that physical rehabilitation is safe, with low adverse-event rates. Furthermore, multiprofessional strategies that span across ICU, ward, and community are required to address complex problems including physical, cognitive, and psychological sequelae of critical illness. Research priorities include detailed reporting of intervention dose (timing, intensity, duration) for both usual care and rehabilitation provided within clinical trials, and development of intervention implementation strategies that enhance uptake and fidelity in routine practice. Rehabilitation is integral to contemporary ICU care, spanning the trajectory of recovery into the community. Within the ICU, it requires interprofessional, experienced healthcare personnel to assess clinical status for safe rehabilitation and to identify an individual’s anticipated recovery trajectory. Standardised intervention reporting and implementation-focussed research are essential to advance evidence and improve outcomes for critically ill patients.
Robot-assisted procedures improve precision and patient outcomes, but setup often requires tedious manual positioning due to a lack of intuitive guidance. Augmented reality (AR) can assist with coarse alignment by providing visual cues in the correct anatomical context. However, most AR systems require additional setup steps to calibrate the head-mounted displays and imaging systems. We propose an AR guidance approach that uses only RGB video to perform spatial calibration with minimal setup effort. We establish a shared coordinate frame between the AR headset and the intraoperative imaging system to enable markerless AR guidance. Inter-device poses are estimated directly from RGB video using a Vision Foundation Model (VFM). To improve calibration accuracy in challenging scenarios, such as partial view overlap and long baselines, we apply multi-frame Umeyama optimization with RANSAC, jointly refining scale, rotation, and translation. This process allows users to align the physical robot with the virtual surgical plan with the necessary accuracy. We evaluate calibration and robot alignment accuracy against an ArUco-based baseline, achieving sub-3 ^∘ rotation and 15 mm mean translation error without requiring fiducial markers. We further assessed needle placement using a virtual guide that conveyed both the target trajectory and the robot’s actuation limits. In all cases, the resulting alignment was within a range that allowed the robot to perform fine adjustments and accurately reach the planned trajectory. Experimental results show that our vision-based calibration enables accurate needle alignment within the robot’s workspace while eliminating setup complexity and fiducial dependence. This demonstrates the feasibility of seamless, markerless AR guidance for robotic procedures via vision-based calibration.
We propose straightforward principles to foster an evaluation-forward operating system that can transform the adoption of clinical artificial intelligence from a leap of faith into a stepwise, trust-building process.
Importance Lymph node (LN) metastasis is a strong predictor of tumor recurrence following pancreatectomy for localized pancreatic neuroendocrine tumors (PanNETs). However, most patients lack LN metastasis and many tumors recur. Tools to guide risk-adapted surveillance in this group are lacking. Objective To develop and validate a tumor recurrence and survival risk score for patients with LN-negative PanNETs. Design, Setting, and Participants This retrospective, case-control study of patients with localized PanNETs took place at 5 high-volume US institutions from 2000 to 2023. Inclusion required 8 or more evaluated LNs and negative nodal status. Median follow-up was 50.6 months. These data were analyzed from March 2025 to May 2025. Exposure Surgical resection of localized PanNETs per clinical guidelines. Main Outcomes and Measures The primary outcome was tumor recurrence. Independent predictors were identified using multivariable logistic regression and used to construct a 13-point composite risk score. Performance was assessed using C statistics. Kaplan-Meier and log-rank methods evaluated disease-free survival (DFS). Genomic profiling was conducted in an external validation cohort to identify and validate recurrence-associated mutational risk scores. Results Of 2024 patients, 770 met inclusion criteria. Median age was 58.7 (IQR, 18.4) years; 405 were male (52.6%) and 365 were female (47.4%). Most tumors were sporadic (94.1%), nonfunctional (90.4%), and located in the body/neck (50.9%). Recurrence occurred in 82 patients (10.6%) at a median of 32.4 (IQR, 16.3-82.0) months after surgery. Independent predictors included male sex (odds ratio [OR], 2.2; 95% CI, 1.3-3.9), tumor size 3 cm or larger (OR, 2.64; 95% CI, 1.5-4.6), World Health Organization grade 2 or higher (OR, 3.70; 95% CI, 1.4-10.0), and lymphovascular invasion (OR, 3.84; 95% CI, 2.1-6.9). The risk score showed strong performance (area under the receiver operating characteristic, 0.83 internally; 0.95 externally). Recurrence rates by risk group were 2.4%, 9.0%, and 27.7% ( P < .001), and 10-year DFS rates of 96.1%, 83.6%, and 51.3%, for low-risk, moderate-risk, and high-risk groups, respectively ( P < .001). Genomic analyses revealed higher tumor mutational burden, somatic mutation count, and somatic mutations in CDC42BPB , DAXX , ERI2 , GALNT9 , MTOR , NUMA1 , and TRPC7 genes among recurrent tumors. Conclusions and Relevance Despite LN-negative status, a subset of patients with PanNETs remained at high risk for recurrence. This validated risk score stratifies recurrence and survival risk showing biological relevance. These findings provide a framework for refining postoperative surveillance and risk-adapted therapeutic strategies.
The growing number of bariatric and oncologic gastrointestinal and hepatopancreatobiliary surgeries has fueled the increased need for biliary interventions in patients with surgically altered anatomy (SAA), where standard endoscopic retrograde cholangiopancreatography (ERCP) is often unfeasible. Traditional alternative methods, such as enteroscopy- or laparoscopy-assisted ERCP, are associated with technical and logistical limitations. Endoscopic ultrasound (EUS)-directed ERCP uses lumen-apposing metal stents (LAMS) to create temporary anastomoses between adjacent lumens, thus restoring access for standard ERCP. These procedures include EUS-directed transgastric (EDGE), transenteric (EDEE), transduodenal (EDDE), and transcholedochal/transcystic (EDCE) ERCP. This review provides an evidence-based analysis of the available techniques and comparative outcomes, with a particular focus on technical and strategic aspects including: (1) how to create an EUS-guided anastomosis across different surgical scenarios; (2) how to perform a through-the-LAMS ERCP, including endoscope and accessory selection, ERCP timing and LAMS fixation; and (3) how to manage the fistula after completion of therapy, including timing of LAMS removal and fistula closure. An algorithm is proposed to guide the endoscopist through the complex decision-making inherent to pancreatobiliary interventions in SAA.
