Same-day discharge following minimally invasive partial and radical nephrectomy: a National Surgical Quality Improvement Program (NSQIP) analysis.

PURPOSE: Minimally invasive partial nephrectomy (MIPN) and radical nephrectomy (MIRN) have successfully resulted in shorter length of stay (LOS) for patients. Using a national cohort, we compared 30-day outcomes of SDD (LOS = 0) versus standard-length discharge (SLD, LOS = 1-3) for MIRN and MIPN. METHODS: All patients who underwent MIPN (CPT 50,543) or MIRN (CPT 50,545) in the ACS-NSQIP database from 2012 to 2019 were reviewed. SDD and SLD groups were matched 1:1 by age, sex, race, body mass index, American Society of Anesthesiologists score, and medical comorbidities. We compared baseline characteristics, 30-day Clavien-Dindo (CD) complications, reoperations, and readmissions between SDD and SLD groups. Multivariable logistic regressions were used to evaluate predictors of adverse outcomes. RESULTS: 28,140 minimally invasive nephrectomy patients were included (SDD n = 237 [0.8%], SLD n = 27,903 [99.2%]). There were no significant differences in 30-day readmissions, CD I/II, CDIII, or CD IV complications before and after matching SDD and SLD groups. On multivariate regression analysis, SDD did not confer increased risk of 30-day complications or readmissions for both MIPN and MIRN. CONCLUSION: SDD after MIPN and MIRN did not confer increased risk of postoperative complications, reoperation, or readmission compared to SLD. Further research should explore optimal patient selection to ensure safe expansion of this initiative.

Adenine nucleotide metabolism and a role for AMP in modulating flagellar waveforms in mouse sperm.

While most ATP, the main energy source driving sperm motility, is derived from glycolysis and oxidative phosphorylation, the metabolic demands of the cell require the efficient use of power stored in high-energy phosphate bonds. In times of high energy consumption, adenylate kinase (AK) scavenges one ATP molecule by transphosphorylation of two molecules of ADP, simultaneously yielding one molecule of AMP as a by-product. Either ATP or ADP supported motility of detergent-modeled cauda epididymal mouse sperm, indicating that flagellar AKs are functional. However, the ensuing flagellar waveforms fueled by ATP or ADP were qualitatively different. Motility driven by ATP was rapid but restricted to the distal region of the sperm tail, whereas ADP produced slower and more fluid waves that propagated down the full flagellum. Characterization of wave patterns by tracing and superimposing the images of the flagella, quantifying the differences using digital image analysis, and computer-assisted sperm analysis revealed differences in the amplitude, periodicity, and propagation of the waves between detergent-modeled sperm treated with either ATP or ADP. Surprisingly, addition of AMP to the incubation medium containing ATP recapitulated the pattern of sperm motility seen with ADP alone. In addition to AK1 and AK2, which we previously demonstrated are present in outer dense fibers and mitochondrial sheath of the mouse sperm tail, we show that another AK, AK8, is present in a third flagellar compartment, the axoneme. These results extend the known regulators of sperm motility to include AMP, which may be operating through an AMP-activated protein kinase.

Comparison of Single-Port Robotic Donor Nephrectomy and Laparoscopic Donor Nephrectomy.

Purpose: To compare the intra- and postoperative outcomes of single-port robotic donor nephrectomies (SP RDNs) and laparoscopic donor nephrectomies (LDNs). Materials and Methods: We retrospectively reviewed our institutional database for patients who received LDN or SP RDN between September 2020 and December 2022. Donor baseline characteristics, intraoperative outcomes, postoperative outcomes, and recipient renal function were extracted and compared between LDN and SP RDN. SP RDN learning curve analysis based on operative time and graft extraction time was performed using cumulative sum analysis. Results: One hundred forty-four patients underwent LDN and 32 patients underwent SP RDN. LDN and SP RDN had similar operative times (LDN: 190.3 ± 28.0 minutes, SP RDN: 194.5 ± 35.1 minutes, p = 0.3253). SP RDN patients had significantly greater extraction times (LDN: 83.2 ± 40.3 seconds, SP RDN: 204.1 ± 52.2 seconds, p < 0.0001) and warm ischemia times (LDN: 145.1 ± 61.7 seconds, SP RDN: 275.4 ± 65.6 seconds, p < 0.0001). There were no differences in patient subjective pain scores, inpatient opioid usage, or Clavien-Dindo II+ complications. Short- and medium-term postoperative donor and recipient renal function were also similar between the groups. SP RDN graft extraction time and total operative time learning curves were achieved at case 27 and 13, respectively. Conclusion: SP RDN is a safe and feasible alternative to LDN that minimizes postoperative abdominal incisional scars and has a short learning curve. Future randomized prospective clinical trials are needed to confirm the findings of this study and to identify other potential benefits and drawbacks of SP RDNs.

Swiss LithoClast® Trilogy Lithotripter for Use in Robotic Pyelolithotomy.

Introduction: The Boston Scientific Swiss LithoClast® Trilogy lithotripter was intended for use in percutaneous nephrolithotomy. We performed, to our knowledge, the first two robotic pyelolithotomies using the Trilogy lithotripter for intracorporeal lithotripsy. Case Description: Two cases are presented involving a 65-year-old female with a complete left staghorn calculus and hydronephrosis secondary to a left ureteropelvic junction (UPJ) obstruction, and a 69-year-old male with a large left staghorn calculus and multiple large left sided simple renal cysts. In both cases, a robotic pyelolithotomy was scheduled for stone removal along with concurrent UPJ repair and cyst decortication respectively. Following pyeloplasty and cyst decortication respectively, and following stone visualization, the 2.4-mm Trilogy probe was inserted into the 12-mm assistant port and under direct visualization the stone was fragmented and removed using Trilogy's built-in mechanisms. Both patients were treated successfully without complications and were found to be stone-free on follow-up. Conclusion: The Trilogy lithotripter may be an effective tool for stone management when introduced during robotic pyelolithotomy and provides additional optionality when manual extraction poses challenges.

A Comparison of Pharmaceutical Prices for Generic Erectile Dysfunction and Benign Prostatic Hyperplasia Medication in 2 Socioeconomically Disparate New York City Neighborhoods.

INTRODUCTION: Erectile dysfunction (ED) and benign prostatic hyperplasia (BPH) are frequently managed with medications. Variability and poor understanding of medication prices have been shown to hinder patient adherence, leading to worse clinical outcomes. We sought to explore how pharmacy type and neighborhood socioeconomic status influence pricing for generic ED and BPH medications. METHODS: A total of 96 pharmacies within the adjacent higher income Upper East Side (UES) and lower income East Harlem (EH) New York City neighborhoods were classified as chain or independent. Telephone surveys identified cash prices for 30-day supplies of 14 medications including phosphodiesterase 5-inhibitors, 5-alpha reductase inhibitors, alpha blockers and antispasmodics. Pricing variability based on pharmacy type and neighborhood was evaluated using Mann-Whitney U-tests. RESULTS: Of 96 pharmacies, 81 responded (84.4%). Independent pharmacies showed significantly reduced prices for 9/14 and 14/14 medications in UES and EH, respectively. The greatest independent pharmacy price reductions were for tadalafil 20 mg (15.0-fold in UES, 26.7-fold in EH) and sildenafil (8.4-fold in UES, 15.4-fold in EH). The least significant reductions were in mirabegron (1.1-fold in UES, 1.2-fold in EH). Independent pharmacies in EH showed lower prices for 9/14 medications compared to those in UES. CONCLUSIONS: Across both neighborhoods, independent pharmacies offered consistently lower cash prices for ED and BPH medications. Lower independent pharmacy prices in the lower income EH neighborhood suggest that neighborhood socioeconomic status may impact pricing. Physicians and patients alike must understand the factors that influence pricing to ensure more optimal patient compliance for uninsured patients.

Adrenalectomy outcomes predicted by a 5-item frailty index (5-iFI) in the ACS-NSQIP database.

BACKGROUND: Frailty has been shown to be a predictor of adverse postoperative outcomes. This study aims to evaluate a 5-item frailty index (5-iFI) as a predictor of complications as well as healthcare resource utilization (HCRU) following adrenalectomy. METHODS: All adrenalectomy cases recorded in the ACS-NSQIP database from 2015 to 2018 were analyzed. Primary outcomes of interest were Clavien-Dindo [CD] I/II or CD IV complications and HCRU. HCRU outcomes were prolonged length of stay (PLOS), discharge to continued care (DCC), and unplanned 30-day readmission (UR). RESULTS: 4358 patients were included. Higher 5-iFI scores were associated with higher rates of CDI/II, CDIV, and increased HCRU (p < 0.05). On multivariate analysis, 5-iFI scores were found to be independent predictors of adverse clinical and HCRU outcomes. CONCLUSIONS: Frailty tools like the 5i-FI can be useful in preoperative risk-benefit analysis, patient counseling, and planning prehabilitation interventions.

The Utilization and Safety of Same-Day Discharge After Transurethral Benign Prostatic Hyperplasia Surgery: A Case-Control, Matched Analysis of a National Cohort.

OBJECTIVE: To analyze the utilization and safety of same-day (SDD) vs standard-length discharge (SLD) for transurethral resection (TURP), holmium laser enucleation (HoLEP), and GreenLight photovaporization (GL-PVP) of the prostate. METHODS: Using the 2015-2019 ACS-NSQIP files, the annual proportion of TURP, HoLEP, and GL-PVP performed with SDD (length of stay [LOS] = 0 days) was calculated. Patients were stratified by LOS into SDD and SLD (TURP: LOS = 1-3 days, HoLEP and GL-PVP: LOS = 1-2 days); those with longer LOS were excluded. Patients were matched 1:1 by age, body mass index, American Society of Anesthesiologists score, and modified Charlson Comorbidity Index score. We compared 30-day unplanned readmissions, reoperations, and Clavien-Dindo (CD) complications between SLD and SDD, and evaluated predictors of adverse outcomes using logistic regression. RESULTS: Most GL-PVP patients underwent SDD, compared to a minority of TURP and HoLEP patients. SDD utilization increased, remained stable, and decreased over time for HoLEP, TURP, and GL-PVP, respectively. For 46,898 included cases (31,872 TURP, 2,901 HoLEP, 12,125 GL-PVP), rates of reoperation, CD I/II, or CD IV complications were comparable before and after matching. Compared to SLD, 30-day unplanned readmission rates for matched SDD patients were lower following TURP (3.48% vs 4.25%, P = .013) and HoLEP (1.93% vs 4.43%, P = .003). On multivariate regression, SLD correlated with unplanned readmission after TURP and HoLEP for both unmatched and matched cohorts. CONCLUSION: For appropriately selected patients, SDD after TURP, HoLEP, and GL-PVP did not confer increased risk of 30-day complications, suggesting patient selection for SDD is being done with appropriate safety nationally.

Robotic Single-Port Donor Nephrectomy with the da Vinci SP® Surgical System.

OBJECTIVES: The da Vinci SP® Surgical System received U.S. Food and Drug Administration approval for urological procedures in 2018. Here, we describe the first experience performing single-port robot-assisted donor nephrectomy (RADN) using the da Vinci SP® surgical system, present 90-day clinical outcomes, and discuss tips for operative success. METHODS: Seven consecutive patients underwent single-port RADN at a single institution between September 1, 2020 and March 31, 2021. Surgery was performed through a single, 60 mm Pfannenstiel incision with a 12 mm periumbilical assistant port for suction and vascular stapling. Donor characteristics, operative details, 90-day donor clinical outcomes, and recipient renal function were retrospectively evaluated. RESULTS: Four female and three male patients successfully underwent single-port RADN without conversion to standard multiport or open approach. Six cases were left-sided. Estimated blood loss for each procedure was ≤ 50 mL. Mean operative time, warm ischemia time, and extraction time were 218.3 minutes (standard deviation [SD]: 16.3 minutes), 5 minutes 4 seconds (SD: 56 seconds), and 3 minutes 37 seconds (SD: 38 seconds). Mean pre-operative creatinine and estimated glomerular filtration rate were 0.79 mg/dL and 107.3 mL/min/1.73m2, respectively. At six week's follow up, they were 1.22 mg/dL and 66.1 mL/min/1.73m2. Average pain score at 48 hours postoperatively was 1.7/10. There were no Clavien-Dindo grade ≥ III complications within 90 days. All recipients experienced immediate and sustained return of renal function post-transplant. CONCLUSION: Single-port RADN is a technically feasible and safe procedure with the da Vinci SP® system and can confer acceptable functional and cosmetic outcomes. Future studies are needed to define long-term outcomes and compare with previously established techniques for donor nephrectomy.

Disposable versus Reusable Cystoscopes: A Micro-Costing Value Analysis in High-Volume and Low-Volume Urology Practices.

INTRODUCTION: Disposable single-use cystoscopes have become increasingly available, demonstrating comparable quality to reusable cystoscopes while eliminating the need for reprocessing and repairs. However, high costs remain a concern. To clarify the role for these scopes, we performed a cost analysis comparison between the single-use Ambu® aScope™ 4 cystoscope and reusable Olympus® CYF-VHR and V2 cystoscopes in 2 clinical settings: a high-volume multi-provider practice and low-volume single-provider practice. METHODS: The number of cystoscopies at each center was recorded between January and December 2019. Elements in the micro-costing analysis included the original purchasing price of the cystoscopes plus accessory equipment, sterilization supplies, repair costs, and personnel. Costs were amortized over 5 or 10 years and calculated on a per-case basis. An annual total cost analysis was performed to evaluate the cost-effectiveness of each device for each facility. RESULTS: In 2019, 1,984 and 245 cystoscopic procedures were performed at the high and low-volume clinics, respectively. At the high-volume multi-provider practice, per-case cost for reusable cystoscopy amounted to $65.98 compared to $227.18 for single-use cystoscopy, with reusable equipment more cost-effective after 294 cystoscopies. At the low-volume single-provider practice, the per-case cost for reusable cystoscopy was $232.62 compared to $461.18 for single-use cystoscopy, with reusable equipment more cost-effective after 19 cases. CONCLUSIONS: Based on this micro-costing analysis, per-case costs favor reusable cystoscopes. While single-use cystoscope pricing may be prohibitive for large and small facilities at this present time, these instruments are powerful adjuncts to urologists' armamentaria when portability and efficiency are prioritized.

Clinically node-positive (cN+) urothelial carcinoma of the bladder treated with chemotherapy and radical cystectomy: Clinical outcomes and development of a postoperative risk stratification model.

INTRODUCTION AND OBJECTIVE: Although node-positive (cN+) bladder cancer is considered Stage IV disease, a subset of patients is treated with chemotherapy and consolidative radical cystectomy (RC). We examined the clinical outcomes of such patients and developed a risk prediction model to facilitate risk-stratification and management. METHODS: We identified adult patients with cTany cN1-3 M0 urothelial carcinoma of the bladder treated with chemotherapy followed by RC from 2006 to 2013 in the NCDB. The associations of clinicopathologic features with overall survival (OS) were evaluated using Cox regression, and a simplified risk score was developed. RESULTS: A total of 491 patients received chemotherapy followed by RC. Median number of lymph nodes removed was 16 (interquartile range 9-25). At RC, 10% of patients were ypT0, and 35% were ypN0. Over a median follow-up of 18.7 months, 160 patients died of any cause. 1-, 5-, and 8-year OS were 69%, 34%, and 29%, respectively. On multivariable analysis, pT stage (hazard ratio [HR] 2.18; P = 0.003 for pT3, HR 2.65; P < 0.001 for pT4 vs.

Indications for stent omission after ureteroscopic lithotripsy defined: A single-institution experience with cost analysis.

Objectives: To report on our experience with the use of an evidence-based algorithm defining specific indications for stent omission (SO) after ureteroscopic lithotripsy (URSL), as stent placement has been associated with increased cost and morbidity and indications for SO in the setting of uncomplicated ureteroscopy have been proposed but remain vague. Patients and methods: Indications for SO were defined as per the attached figure, data from URSL procedures performed from January 2016 to September 2017 were collected. For procedures eligible for SO, preoperative and intraoperative factors were recorded including: stone burden, presence of preoperative stent, procedure time, access sheath use, and whether SO was performed. Morbidity data were reviewed including: postoperative events, patient telephone calls for bothersome symptoms, unplanned return visits, and admissions within 30 days. Results: In all, 250 URSL procedures were performed during the study period, and 106 (42.4%) were eligible for SO. SO was performed in 60 (24.0%) cases reflecting a 56.7% compliance with the algorithm. There were no readmissions or re-operations within 30 days for the SO group. Lower postoperative event rates were noted in the SO group (16.7% vs 34.8%, P = 0.03), unplanned return visits (8.3% vs 17.4%, P = 0.16) and 30-day readmission rates (0.0% vs 6.5%, P = 0.08) were also lower in the SO group, although they did not reach statistical significance. Analysis also demonstrated a protective effect of SO on unplanned return visits (odds ratio 0.43, 95% confidence interval 0.13-1.42, P = 0.17), although this was not statistically significant. No statistically significant associations were noted between postoperative events and stone burden, procedure time, or presence of preoperative stent. Conclusions: We provide an algorithm defining indications for SO. SO is safe in a significant portion of URSL procedures, and SO appears to decrease postoperative events when performed judiciously. Abbreviations: IQR: interquartile range; LUTS: lower urinary tract symptoms; OR, odds ratio; SO: stent omission; URSL: ureteroscopic lithotripsy; YAG: yttrium-aluminium-garnet.

Adult-type granulosa cell tumour of the testis: Report of a case and review of the literature.

Granulosa cell tumours (GCTs) can be either juvenile or adult type, and more commonly occur in the ovaries. Adult-type GCTs of the testis (AGCTT) are very rare and only 46 cases have previously been reported. We report here on a 48-year-old Filipino man with a left testicular AGCTT, which measured 1.2 × 1.2 × 1.0 cm. He underwent radical orchidectomy with postoperative surveillance for 1 year, which included computed tomography with oral intravenous contrast and clinical examinations, which have been unremarkable. The previously reported AGCTTs were briefly reviewed.

Endoscopic scene labelling and augmentation using intraoperative pulsatile motion and colour appearance cues with preoperative anatomical priors.

PURPOSE: Despite great advances in medical image segmentation, the accurate and automatic segmentation of endoscopic scenes remains a challenging problem. Two important aspects have to be considered in segmenting an endoscopic scene: (1) noise and clutter due to light reflection and smoke from cutting tissue, and (2) structure occlusion (e.g. vessels occluded by fat, or endophytic tumours occluded by healthy kidney tissue). METHODS: In this paper, we propose a variational technique to augment a surgeon's endoscopic view by segmenting visible as well as occluded structures in the intraoperative endoscopic view. Our method estimates the 3D pose and deformation of anatomical structures segmented from 3D preoperative data in order to align to and segment corresponding structures in 2D intraoperative endoscopic views. Our preoperative to intraoperative alignment is driven by, first, spatio-temporal, signal processing based vessel pulsation cues and, second, machine learning based analysis of colour and textural visual cues. To our knowledge, this is the first work that utilizes vascular pulsation cues for guiding preoperative to intraoperative registration. In addition, we incorporate a tissue-specific (i.e. heterogeneous) physically based deformation model into our framework to cope with the non-rigid deformation of structures that occurs during the intervention. RESULTS: We validated the utility of our technique on fifteen challenging clinical cases with 45 % improvements in accuracy compared to the state-of-the-art method. CONCLUSIONS: A new technique for localizing both visible and occluded structures in an endoscopic view was proposed and tested. This method leverages both preoperative data, as a source of patient-specific prior knowledge, as well as vasculature pulsation and endoscopic visual cues in order to accurately segment the highly noisy and cluttered environment of an endoscopic video. Our results on in vivo clinical cases of partial nephrectomy illustrate the potential of the proposed framework for augmented reality applications in minimally invasive surgeries.

Simultaneous Multi-Structure Segmentation and 3D Nonrigid Pose Estimation in Image-Guided Robotic Surgery.

In image-guided robotic surgery, segmenting the endoscopic video stream into meaningful parts provides important contextual information that surgeons can exploit to enhance their perception of the surgical scene. This information provides surgeons with real-time decision-making guidance before initiating critical tasks such as tissue cutting. Segmenting endoscopic video is a challenging problem due to a variety of complications including significant noise attributed to bleeding and smoke from cutting, poor appearance contrast between different tissue types, occluding surgical tools, and limited visibility of the objects' geometries on the projected camera views. In this paper, we propose a multi-modal approach to segmentation where preoperative 3D computed tomography scans and intraoperative stereo-endoscopic video data are jointly analyzed. The idea is to segment multiple poorly visible structures in the stereo/multichannel endoscopic videos by fusing reliable prior knowledge captured from the preoperative 3D scans. More specifically, we estimate and track the pose of the preoperative models in 3D and consider the models' non-rigid deformations to match with corresponding visual cues in multi-channel endoscopic video and segment the objects of interest. Further, contrary to most augmented reality frameworks in endoscopic surgery that assume known camera parameters, an assumption that is often violated during surgery due to non-optimal camera calibration and changes in camera focus/zoom, our method embeds these parameters into the optimization hence correcting the calibration parameters within the segmentation process. We evaluate our technique on synthetic data, ex vivo lamb kidney datasets, and in vivo clinical partial nephrectomy surgery with results demonstrating high accuracy and robustness.

Automatic segmentation of occluded vasculature via pulsatile motion analysis in endoscopic robot-assisted partial nephrectomy video.

Hilar dissection is an important and delicate stage in partial nephrectomy, during which surgeons remove connective tissue surrounding renal vasculature. Serious complications arise when the occluded blood vessels, concealed by fat, are missed in the endoscopic view and as a result are not appropriately clamped. Such complications may include catastrophic blood loss from internal bleeding and associated occlusion of the surgical view during the excision of the cancerous mass (due to heavy bleeding), both of which may compromise the visibility of surgical margins or even result in a conversion from a minimally invasive to an open intervention. To aid in vessel discovery, we propose a novel automatic method to segment occluded vasculature from labeling minute pulsatile motion that is otherwise imperceptible with the naked eye. Our segmentation technique extracts subtle tissue motions using a technique adapted from phase-based video magnification, in which we measure motion from periodic changes in local phase information albeit for labeling rather than magnification. Based on measuring local phase through spatial decomposition of each frame of the endoscopic video using complex wavelet pairs, our approach assigns segmentation labels by detecting regions exhibiting temporal local phase changes matching the heart rate. We demonstrate how our technique is a practical solution for time-critical surgical applications by presenting quantitative and qualitative performance evaluations of our vessel detection algorithms with a retrospective study of fifteen clinical robot-assisted partial nephrectomies.

Auto localization and segmentation of occluded vessels in robot-assisted partial nephrectomy.

Hilar dissection is an important and delicate stage in partial nephrectomy during which surgeons remove connective tissue surrounding renal vasculature. Potentially serious complications arise when vessels occluded by fat are missed in the endoscopic view and are not appropriately clamped. To aid in vessel discovery, we propose an automatic method to localize and label occluded vasculature. Our segmentation technique is adapted from phase-based video magnification, in which we measure subtle motion from periodic changes in local phase information albeit for labeling rather than magnification. We measure local phase through spatial decomposition of each frame of the endoscopic video using complex wavelet pairs. We then assign segmentation labels based on identifying responses of regions exhibiting temporal local phase changes matching the heart rate frequency. Our method is evaluated with a retrospective study of eight real robot-assisted partial nephrectomies demonstrating utility for surgical guidance that could potentially reduce operation times and complication rates.

Efficient multi-organ segmentation in multi-view endoscopic videos using pre-operative priors.

Synergistic fusion of pre-operative (pre-op) and intraoperative (intra-op) imaging data provides surgeons with invaluable insightful information that can improve their decision-making during minimally invasive robotic surgery. In this paper, we propose an efficient technique to segment multiple objects in intra-op multi-view endoscopic videos based on priors captured from pre-op data. Our approach leverages information from 3D pre-op data into the analysis of visual cues in the 2D intra-op data by formulating the problem as one of finding the 3D pose and non-rigid deformations of tissue models driven by features from 2D images. We present a closed-form solution for our formulation and demonstrate how it allows for the inclusion of laparoscopic camera motion model. Our efficient method runs in real-time on a single core CPU making it practical even for robotic surgery systems with limited computational resources. We validate the utility of our technique on ex vivo data as well as in vivo clinical data from laparoscopic partial nephrectomy surgery and demonstrate its robustness in segmenting stereo endoscopic videos.

Automatic labelling of tumourous frames in free-hand laparoscopic ultrasound video.

Laparoscopic ultrasound (US) is often used during partial nephrectomy surgeries to identify tumour boundaries within the kidney. However, visual identification is challenging as tumour appearance varies across patients and US images exhibit significant noise levels. To address these challenges, we present the first fully automatic method for detecting the presence of kidney tumour in free-hand laparoscopic ultrasound sequences in near real-time. Our novel approach predicts the probability that a frame contains tumourous tissue using random forests and encodes this probability combined with a regularization term within a graph. Using Dijkstra's algorithm we find a globally optimal labelling (tumour vs. non-tumour) of each frame. We validate our method on a challenging clinical dataset composed of five patients, with a total of 2025 2D ultrasound frames, and demonstrate the ability to detect the presence of kidney tumour with a sensitivity and specificity of 0.774 and 0.916, respectively.