BACKGROUND:Vacuum-assisted breast biopsy (VABB) enables minimally invasive sampling of breast lesions but provides no real-time information about the tissue being sampled, potentially leading to sampling bias. We evaluated whether integrating diffuse reflectance spectroscopy (DRS) into the VABB workflow is feasible and can support more targeted sampling. METHODS:A fiber-equipped introducer compatible with a standard Hologic VABB system was developed to acquire DRS measurements without altering the clinical workflow. In 40 ex vivo mastectomy specimens, ultrasound-guided VABB was performed, with DRS measurements obtained immediately before biopsy acquisition. Histopathology served as the reference standard, and labeled spectra were used to train machine learning models using sample-wise and patient-wise classification to distinguish healthy from malignant tissue. RESULTS:A total of 300 biopsies were correlated with histopathology, yielding a balanced dataset of 248 spectra. The sample-wise model achieved a mean accuracy of 86%, sensitivity of 91%, and specificity of 82%. The patient-wise approach achieved 76%, 85%, and 70%, respectively. CONCLUSIONS:This proof-of-concept study demonstrates that DRS can be integrated into the VABB workflow using a novel optical introducer and discriminate malignant from healthy tissue in ex vivo breast specimens, supporting technical feasibility. Further clinical studies are needed to establish its diagnostic value.
BACKGROUND:We recently developed a novel flexible suction/coagulation probe to overcome the rigidity of conventional electrosurgical probes, enabling simultaneous suction and coagulation from the console in robotic surgery. Although feasible in robot-assisted gastrectomy, its performance remains unevaluated in robot-assisted hepatectomy, which requires frequent suction and reliable hemostasis. METHODS:This single-arm feasibility study included 12 consecutive patients undergoing robot-assisted hepatectomy using the probe, directly manipulated by the console surgeon for hemostasis during parenchymal transection. Surgical outcomes (operative time, estimated blood loss, device trouble, complications) were assessed. Surgeons completed a post-procedure questionnaire covering seven usability and performance domains on a 3-point Likert scale. RESULTS:Robotic procedures included left hepatectomy, segmentectomies, and partial resections. All procedures were completed successfully without conversion. Median operative time was 397.5 min (range, 256-678 min), and estimated blood loss was 110 mL (range, 0-320 mL). No device-related adverse events or complications occurred. Overall satisfaction was rated 'satisfactory' by 75% of surgeons, with 'unsatisfactory' ratings occurring only for suction efficiency (16.7%). CONCLUSIONS:The flexible suction/coagulation probe appears to be a practical and clinically feasible option for robot-assisted hepatectomy. Although the suction efficiency requires further improvement, the device enables direct console surgeon-controlled suction and hemostasis during hepatic parenchymal transection.
BACKGROUND:Artificial Intelligence (AI) is rapidly transitioning from experimental research to daily medical practice, yet the medical community's understanding of these tools remains largely confined to visible 'front-end' applications with which the clinician can directly interact (such as decision support systems and conversational agents). This perspective overlooks the proliferation of 'back-end' AI, the algorithms that are embedded within devices, systems, and hospital infrastructure that silently reconstruct data while remaining invisible to clinicians. This paper presents a clinician-oriented framework that distinguishes between these two categories and proposes a clinically oriented guide for their evaluation. METHODS:We performed a narrative review of AI applications in minimally invasive therapy and medical imaging to categorize systems into 'front-end' (interactive/visible) and 'back-end' (embedded/invisible) modalities. We synthesized evaluation metrics from computer science and engineering literature, selecting those relevant for clinical safety and decision-making to create a practical literacy guide. RESULTS:Front-end and back-end systems require distinct validation strategies to ensure safety. while front-end evaluation must prioritize decision quality, spatial precision, and human-computer interaction to mitigate risks like automation bias, back-end evaluation requires rigorous technical benchmarking of signal fidelity and temporal latency to ensure that algorithmic reconstruction does not distort clinical reality. To facilitate this, we developed a structured inquiry framework to guide clinicians in auditing these systems for data provenance, transparency, and failure modes. Crucially, we emphasize that mathematical optimization does not guarantee clinical efficacy; technical metrics must always be paired with specific clinical contexts to ensure they align with patient-centered outcomes. CONCLUSION:Clinical safety in the AI era demands 'algorithmic literacy'. By applying this front-end/back-end framework and understanding key technical metrics, medical professionals can better identify failure modes, ensure data integrity, and maintain clear lines of clinical accountability, shifting from passive consumers to active evaluators of medical technology.
BACKGROUND:Suturing complex anastomoses requires repetitive training for mastery. Expert tutoring is a limited and costly ressource. This study assessed the added value of surgical video instructions in expert-tutored hepatopancreatobiliary (HPB) surgery courses. METHODS:Trainees and tutors used QR codes to access instructional videos of pancreatico- and hepaticojejunostomy (04/2024-09/2025: Davos course for HPB Surgery; Hands-On- Course of the North-German Surgical Association; ESSO Course Pancreatic Surgery). After watching videos and performing anastomoses on 3D-biotissue organ models, participants anonymously completed standardized questionnaires. RESULTS:Out of 51 participants (20 residents; 31 specialists), most reported that QR codes facilitated video access (n=48, 96%). The majority found the videos useful (n=31, 62%). Most preferred a combination of expert tutoring and surgical video instructions (n=49, 96%) and agreed that expert tutoring was more effective when supported by videos (n=44, 86%). Furthermore, 28 (55%) participants felt they could improve their surgical skills with the videos in self-directed training without expert guidance. Older and more experienced participants were more likely to agree that videos added value to expert tutoring. CONCLUSION:Surgical video instructions were well accepted and perceived as valuable additions to expert-tutored courses on HPB anastomoses. Further studies should assess translation into improved training outcomes.
BACKGROUND:In flexible endoscopy, ergonomic issues such as suboptimal monitor positioning are becoming increasingly important. Head-mounted display (HMD) technology may provide an ergonomic environment independent of room setup or endoscopist posture. This study evaluated the feasibility and ergonomic benefits of immersive endoscopy using HMD technology. METHODS:We used the Apple Vision Pro (AVP), a video see-through HMD. Participants performed an endoscopic marking task under three monitor settings: ergonomic, non-ergonomic, and immersive. Procedure time, muscle effort by wireless electromyography, and mental workload by NASA-TLX were compared. RESULTS:Eleven endoscopists participated. Compared with the non-ergonomic setting, the immersive and ergonomic settings showed significantly shorter procedure times and lower mental workload, with no difference between the immersive and ergonomic settings. Both immersive and ergonomic settings also showed significantly lower muscle activation in the trapezius, posterior neck, and left sternocleidomastoid muscles than the non-ergonomic setting. In addition, compared with the ergonomic setting, immersive endoscopy showed significantly lower activation in the trapezius, erector spinae, and right sternocleidomastoid muscles. CONCLUSIONS:Immersive endoscopy using AVP improved performance and reduced muscle effort and mental workload compared with the non-ergonomic setting, while reducing activation in specific postural muscles even compared with the ergonomic setting.
BACKGROUND:This study aimed to investigate and compare the postoperative effects of vaginal natural orifice transluminal endoscopic surgery (vNOTES) and conventional laparoscopy on patients' sexual function, quality of life, and self-esteem for benign indications. METHODS:Data from patients who underwent hysterectomies and tubal surgeries using either vNOTES or conventional laparoscopy between January 2022 and August 2023 were analyzed and compared. The primary outcomes were sexual function, quality of life, and self-esteem, assessed both preoperatively and at 3 months post-surgery. Other outcomes of interest, such as surgical success, duration of surgery, complication rates, and postoperative pain, were also included. RESULTS:Among the women included in the study, 105 (34.8%) underwent total laparoscopic hysterectomy (TLH), 63 (20.9%) underwent vNOTES hysterectomy (VH), 53 (17.5%) underwent laparoscopic tubal surgery (LTS), and 81 (26.8%) underwent vNOTES tubal surgery (VTS). Postoperative Short Form-12 (SF-12) Physical Component Summary (PCS) scores improved significantly in the hysterectomy groups (TLH: 41.16 ± 4.85; VH: 39.68 ± 4.73; p = 0.001). Similarly, Mental Component Summary (MCS) scores also improved postoperatively (TLH: 42.54 ± 5.05; VH: 39.68 ± 4.26; p = 0.001). In the tubal surgery groups, postoperative PCS scores were 42.87 ± 3.38 for LTS and 41.6 ± 3.25 for VTS, with significant improvements observed in both LTS (p = 0.033) and VTS (p = 0.002). However, postoperative MCS changes were not statistically significant for either LTS (p = 0.808) or VTS (p = 0.072). Postoperative Female Sexual Function Index (FSFI) scores increased significantly in both hysterectomy groups (TLH: 20.73 ± 6.26; VH: 21.52 ± 5.23; p = 0.001). Preoperative FSFI scores were similar between the LTS (23.67 ± 6.88) and VTS (23.75 ± 3.26) groups (p = 0.741). Postoperatively, mean FSFI scores increased to 26.98 ± 5.35 in the LTS group and 29.95 ± 4.31 in the VTS group, demonstrating significant improvement (p = 0.001). On the Rosenberg Self-Esteem Scale, only the VH group showed a significant postoperative improvement (p = 0.001). CONCLUSIONS:The vNOTES procedures yielded results comparable to those of the control group while demonstrating notable improvements in patient-centered outcomes, including sexual function, quality of life, and self-esteem, alongside favorable perioperative surgical outcomes.
PurposeTo evaluate the feasibility of indocyanine green (ICG) fluorescence as an intraoperative tool for abdominal lymph-node localization in cases where fine needle aspiration cytology (FNAC), is not feasible. ICG was administered either 24 hours before surgery or at anaesthesia induction, and fluorescence findings were descriptively compared with preoperative imaging and histology.Materials and methodsThis is a retrospective exploratory feasibility study. Patients undergoing excision of primary or metastatic lesions in areas not suitable for ultrasound-guided biopsy were included. Suspicious findings on computed tomography (CT) scan were confirmed with positron emission tomography (PET)-CT preoperatively. According to intravenous ICG administration, patients were divided into two groups: group A, 24-hours before surgery, and group B, at anaesthesia induction. Fluorescence visualization was descriptively assessed in relation to PET-CT findings and histology.ResultsEight patients were included in group A and nine in group B. ICG fluorescence enabled intraoperative visualization of 13 out of 17 lesions overall (76.5%), including six out of eight in group A (75%), and seven out of nine in group B (77.8%), including retroperitoneal, mesenteric, and supraclavicular locations. Because PET-CT positivity was part of the inclusion pathway, descriptive concordance indicators are reported for descriptive purposes only. Visualization rates were comparable between the two protocols.ConclusionsFluorescence appears to be a feasible and safe intraoperative adjunct for lymph-node identification in anatomically challenging regions. Administration at anaesthesia induction simplifies perioperative workflow while maintaining similar visualization rates. These findings are exploratory and hypothesis-generating and should be interpreted considering the small sample size and the selection bias related to PET-positive inclusion criteria.
BACKGROUND:This study aimed to evaluate the clinical feasibility, usability, and cognitive workload of a certified mixed reality (MXR) platform for preoperative planning and intraoperative navigation in liver surgery. Curative treatment of liver tumors requires precise comprehension of spatial relationships between tumors and vascular structures to achieve complete resection or ablation. Traditional two-dimensional imaging demands cognitive reconstruction of anatomy, posing challenges even for experienced surgeons. MXR technologies, such as Microsoft HoloLens 2 combined with HoloCare Studio™, enable immersive and interactive visualization that may improve spatial orientation, reduce cognitive load and support intraoperative decision-making. METHODS:Routine quality assurance data from five male patients (mean age 61 years; range 44-80), collected during the clinical introduction of a CE-certified MXR device, undergoing open liver resection for hepatic malignancies were retrospectively evaluated. Patient-specific imaging, including three-phase computed tomography (CT) and, in one case, magnetic resonance imaging (MRI), was processed using HoloCare Studio™ (HoloCare Oslo) to create interactive 3D holograms of liver parenchyma, tumors, and vascular structures. These holograms were employed for both preoperative planning and intraoperative navigation through the HoloLens 2 headset. Cognitive workload was assessed using the NASA Task Load Index (NASA-TLX) and usability was measured using the System Usability Scale (SUS). RESULTS:The HoloCare™ MXR system was seamlessly integrated into all five cases and improved tumor localization and understanding of vascular anatomy. Average NASA-TLX scores were 13.8 preoperatively and 14.7 intraoperatively, indicating low cognitive workload. Mean SUS scores were 96.0 preoperatively and 95.5 postoperatively, reflecting excellent usability. CONCLUSION:MXR integration into open hepatic surgery is feasible, fast, imposes minimal cognitive workload, and is well accepted. These findings support further evaluation of MXR as a valuable adjunct in hepatobiliary surgery.
BACKGROUND:This pilot study aimed to assess the feasibility, perioperative safety, and short-term outcomes of extraperitoneal versus transperitoneal vNOTES sacrocolpopexy. METHODS:Thirty-six patients with POP-Q stage III or higher who underwent simultaneous vNOTES hysterectomy, BSO, and sacrocolpopexy were retrospectively analyzed and categorized into extraperitoneal (Group A, n=14) and transperitoneal (Group B, n=22) groups. Anatomical success, quality of life, and recovery were evaluated using POP-Q, P-QOL, ICIQ-UI SF, QoR-15, and MESH-R scales. RESULTS:Baseline characteristics were comparable between groups. Six-month anatomical success rates were similar (extraperitoneal 90.8% vs. transperitoneal 91.6%). The most notable procedural difference was Trendelenburg positioning: Group A was maintained at 10°-15° throughout, whereas Group B required 25°-30° during the sacrocolpopexy phase. Numerical trends favoring the extraperitoneal approach were observed in operative time (99.6 vs. 110.4 min; p=0.185) and time to first stool (27.7 vs. 34.5 h; p=0.552), but neither reached statistical significance. No rectal injuries or intraabdominal abscesses occurred in either group. CONCLUSIONS:Extraperitoneal vNOTES sacrocolpopexy is feasible and safe, yielding anatomical outcomes equivalent to the transperitoneal approach at six months. Its most substantiated advantage is the reduced Trendelenburg angle, which may lower physiological burden in high-BMI patients. Remaining potential benefits require confirmation through larger prospective trials.
BACKGROUND:A deep understanding of anatomy is essential for the professional education of competent clinicians, especially surgeons. This study explores an innovative format of live surgery on body donors complemented by interactive digital technologies and assesses its perceived educational value among students. By evaluating the individual digital technologies, this study also evaluates the format's practical feasibility as a testing platform for digital technologies in surgical anatomy education. METHODS:Operations were performed on body donors by experts in a live format in front of an audience of 40-80 medical students. Surgeons and anatomy specialists interacted with students and explained procedures step by step. Students actively participated in the operations, gaining direct practical experience. In addition, the latest digital interactive technologies, such as virtual reality (VR) and augmented reality (AR), were continuously used and evaluated. RESULTS:Evaluations demonstrated that 96% of 412 participants were 'very enthusiastic' about the concept and appreciated the use of multimedia teaching methods. The Touch Surgery app was well received (mean 3.2 ± SD 1.3; Likert scale 1-5, where 5 indicated 'fully applies'). By using interactive platforms such as Slido and TED, the contents could be repeated and prepared, which was highly appreciated (mean 4.4 ± SD 1.1; Likert scale 1-5, where 5 indicated 'fully applies'). In particular, the use of computed tomography (CT) scans of the body donors was rated positively (mean 3.6 ± SD 1.4; Likert scale 1-5, where 5 indicated 'fully applies'). CONCLUSIONS:Live surgery using body donors improved not only the students' self-reported anatomical knowledge but also the innovative digital technologies with a focus on VR and, in particular, the complementary use of CT datasets of body donors, were perceived positively. This new format is a valuable addition to the standard medical curriculum, providing students with further tools for a better understanding of surgical anatomy and potentially supporting greater confidence in long-term clinical practice.
Background: There are several biliary drainage procedures for biliary strictures, including endoscopic ultrasound-guided hepaticoduodenostomy (EUS-HDS). This multicenter retrospective study evaluated the safety and clinical outcomes of EUS-HDS for intrahepatic bile duct drainage. Methods: Consecutive patients who underwent EUS-HDS at 12 Japanese referral centers between January 2010 and December 2024 were enrolled. The primary endpoint was clinical success. Secondary endpoints were technical success, stent patency, and adverse events. Results: Thirty-five patients were analyzed. Perihilar biliary stenosis was observed in 91.4%, and right posterior sectoral bile ducts were targeted in 80.0%. Technical and clinical success rates were 88.6% and 68.6%, respectively. The cumulative incidence of stent occlusion was 6.7% at 30 days, 13.3% at 90 days, and 20.0% at 180 days. One case of mild peritonitis occurred. Conclusions: EUS-HDS appears to be a feasible and safe option for selected patients with malignant intrahepatic biliary obstruction and may provide clinical benefit as part of a complementary drainage strategy.
BACKGROUND:At present, traditional minimally invasive thermal ablation technologies are all derived from electromagnetic radiation energy generated by high-frequency, high-energy physical sources. Consequently, they all share common issues, such as poor conformability, carbonization, difficulty controlling temperature, and electromagnetic radiation. METHODS:This paper proposes a steam thermal ablation (STA) technology that utilizes the internal energy of saturated steam to replace traditional electromagnetic radiation-based energy sources. Using a self-built, minimally invasive steam thermal ablation precision treatment system for liver tumors and a steam ablation needle designed via COMSOL simulation, ex vivo pig liver experiments were conducted. Based on STA characteristics, a real-time monitoring scheme for the ablation boundary has been proposed using fluorescence imaging. RESULTS:The results of STA are characterized by a maximal ablation axis ratio (short diameter/long diameter) and the absence of carbonization. The fluorescence-based monitoring effectively eliminated artifacts and radiation constraints typical of ultrasound or CT-guided procedures. CONCLUSIONS:Steam thermal ablation (STA) technology offers a novel thermal ablation method, featuring good conformability and no carbonization. Combined with real-time fluorescence imaging monitoring, it provides a feasible solution for the limitations of traditional minimally invasive thermal ablation techniques.
BACKGROUND:This study evaluated the feasibility of lipiodol marking using the left radial artery (LRA) approach prior to cryoablation for small renal cell carcinoma (RCC). METHODS:This retrospective study included 59 patients who underwent transarterial lipiodol marking before cryoablation between August 2022 and December 2023. The LRA approach was attempted in 46 patients; 13 patients ineligible for LRA underwent femoral artery (FA) puncture as a descriptive comparator. The primary endpoint was technical success without crossover among attempted LRA cases; secondary endpoints included overall feasibility among candidates, failure reasons, adverse events (AEs), and procedural metrics. RESULTS:Technical success without crossover was 91.3% (42/46), with overall feasibility of 71.2% (42/59). All four LRA failures resulted from arterial anatomical variation/tortuosity (n = 3) or brachial artery dissection (n = 1), completed via FA crossover. Grade I AEs included forearm haematoma (26.1%), LRA occlusion (2.2%), and brachial artery dissection (2.2%). Procedure and fluoroscopy times were longer in the successful LRA cases than in the FA cohort (median 86 vs. 63 min and 1256 vs. 851 s, respectively). CONCLUSIONS:LRA-based lipiodol marking achieved technical success in most attempted cases. AEs were manageable but not negligible. Prospective studies are warranted to clarify patient-centred benefits and optimal selection criteria.
BACKGROUND:We aimed to evaluate diagnostic performance of preoperative hydrodissection and grey-scale imaging (US) in predicting extra-thyroidal extension (ETE) of subcapsular thyroid lesions. METHODS:The retrospective study evaluated hydrodissection between September 2023 and March 2025 for subcapsular thyroid nodules. Biopsy specimen with atypia of undetermined significance and thyroid tumors without surgery were excluded. For US, ETE was determined based on following features: capsule disruption, bulging contour, and perithyroidal infiltrations. For hydrodissection, ETE was determined positive if resistance was encountered. Diagnostic performance comparing US and hydrodissection in predicting ETE was performed for K-TIRADS 5 nodules and thyroid tumors. RESULTS:A total of 40 nodules per 33 patients (age: 50.2 ± 12.7; female: 28/33, 84.9%) were included. Out of 40 nodules, 18 nodules were tumors, of which 15 nodules (83.3%) were malignant. Four malignant nodules demonstrated ETE. For K-TIRADS 5 nodules (n = 28), hydrodissection demonstrated better diagnostic performance in terms of specificity (0.96 vs. 0.63, p = 0.008), accuracy (0.93 vs. 0.61, p = 0.004), and AUC (0.854 vs. 0.563, p = 0.013). For thyroid tumors (n = 18), hydrodissection demonstrated higher accuracy (0.89 vs. 0.56, p = 0.031) and higher AUC (0.839 vs. 0.536, p = 0.016). CONCLUSION:Hydrodissection demonstrated better specificity, better accuracy, and higher AUC in predicting ETE than US.
BACKGROUND:To evaluate the implementation and clinical impact of a standardised workflow for laparoscopic right hemicolectomy (LRH) with complete mesocolon excision (CME) using a cranial approach in colon cancer, the so-called 'Marginal Gains' project. METHODS:This is a single-centre retrospective before-and-after cohort study of prospectively collected data. Outcomes of 29 patients undergoing LRH from 2022 to 2023 (control group) were compared with 34 patients treated between 2023 and 2024, after implementation of a structured CME-oriented workflow (intervention group). The primary outcome was the rate of workflow application and its clinical impact (secondary outcomes) in terms of CME rate (performed when clinically indicated), intraoperative events, 30-day postoperative complications, and lymph nodes harvested. RESULTS:Baseline characteristics were comparable between groups. Workflow was applied in all cases. CME, performed according to predefined clinical criteria, was more frequent in the intervention group (13/34, 38.2%) than in the control group (1/29, 3.4%; p = .001), reflecting structured workflow adoption and potentially introducing selection bias. Overall postoperative complications occurred in 6/34 (17.6%) patients in the intervention group and 13/29 (44.8%) in the control group (p = .027), mainly driven by low-grade (Clavien I-II) events, particularly paralytic ileus (0 vs. 4 cases) and minor anastomotic bleeding (0 vs. 4 cases). One anastomotic leak occurred in the intervention group (2.9%), requiring reoperation with redo anastomosis (Clavien III-b). No significant differences were observed in tumour stage. The mean number of harvested lymph nodes was higher in the intervention group (22.18 ± 6 vs. 18.83 ± 6.50; p = .021). CONCLUSIONS:Implementation of the standardised CME-oriented workflow was feasible and associated with lower short-term postoperative morbidity and higher lymph node harvest. Given the non-randomised before-and-after design and limited sample size, these findings should be considered exploratory and hypothesis-generating rather than evidence of definitive improvement in surgical or oncological outcomes. Prospective multicentre validation is warranted.
BACKGROUND:Emergency ventral hernia repair remains a challenging procedure due to patient instability, contaminated surgical fields, and heterogeneity in hernia types and operative techniques. Predicting postoperative complications in this setting is difficult using traditional statistical methods. Machine learning (ML) may offer improved predictive accuracy by recognizing nonlinear patterns among multiple perioperative factors. METHODS:A retrospective multicenter analysis was performed using data from the ACTIVE (Acute Treatment for Incisional Ventral Hernias) study, including 557 adult patients undergoing emergent ventral hernia repair between 2018 and 2021 in 31 Italian surgical centers. Demographic, preoperative, intraoperative, and postoperative variables were analyzed. Three ML algorithms-Decision Tree, Random Forest, and Deep Learning Neural Network-were trained and validated using five-fold cross-validation after class balancing with SMOTE. Model performance was compared with traditional logistic regression using accuracy, area under the ROC curve (AUC), and F1 score. RESULTS:Postoperative complications occurred in 181 patients (32.5%), while major complications (Clavien-Dindo ≥ II) occurred in 10%. Random Forest achieved the best performance (AUC 0.95, accuracy 0.88, F1 score 0.86), outperforming logistic regression (AUC 0.82, accuracy 0.78). The most influential predictors were operative duration, ASA score, and sepsis for overall complications, while bowel obstruction and BMI were key factors for major complications. Surgical approach (open vs. laparoscopic) did not independently correlate with adverse outcomes, highlighting the complexity of patient- and case-specific interactions. CONCLUSIONS:Machine learning models can accurately predict postoperative complications following emergent ventral hernia repair, surpassing traditional regression methods. These findings suggest that ML-based decision tools could support risk stratification and optimize surgical planning in high-risk emergency settings. Prospective validation is warranted to integrate AI-assisted prediction into perioperative clinical workflows.
