Corrigendum to "RPN (Radius, Position of tumour, iNvasion of renal sinus) Classification and Nephrometry Scoring System: An Internationally Developed Clinical Classification To Describe the Surgical Difficulty for Renal Masses for Which Robotic Partial Nephrectomy Is Planned" [Eur. Urol. Open Sci. 54 (2023) 33-42].

[This corrects the article DOI: 10.1016/j.euros.2023.05.007.].

RPN (Radius, Position of tumour, iNvasion of renal sinus) Classification and Nephrometry Scoring System: An Internationally Developed Clinical Classification To Describe the Surgical Difficulty for Renal Masses for Which Robotic Partial Nephrectomy Is Planned.

Background: The surgical difficulty of partial nephrectomy (PN) varies depending on the operative approach. Existing nephrometry classifications for assessment of surgical difficulty are not specific to the robotic approach. Objective: To develop an international robotic-specific classification of renal masses for preoperative assessment of surgical difficulty of robotic PN. Design setting and participants: The RPN classification (Radius, Position of tumour, iNvasion of renal sinus) considers three parameters: tumour size, tumour position, and invasion of the renal sinus. In an international survey, 45 experienced robotic surgeons independently reviewed de-identified computed tomography images of 144 patients with renal tumours to assess surgical difficulty of robot-assisted PN using a 10-point Likert scale. A separate data set of 248 patients was used for external validation. Outcome measurements and statistical analysis: Multiple linear regression was conducted and a risk score was developed after rounding the regression coefficients. The RPN classification was correlated with the surgical difficulty score derived from the international survey. External validation was performed using a retrospective cohort of 248 patients. RPN classification was also compared with the RENAL (Radius; Exophytic/endophytic; Nearness; Anterior/posterior; Location), PADUA (Preoperative Aspects and Dimensions Used for Anatomic), and SPARE (Simplified PADUA REnal) scoring systems. Results and limitation: The median tumour size was 38 mm (interquartile range 27-49). The majority (81%) of renal tumours were peripheral, followed by hilar (12%) and central (7.6%) locations. Noninvasive and semi-invasive tumours accounted for 37% each, and 26% of the tumours were invasive. The mean surgical difficulty score was 5.2 (standard deviation 1.9). Linear regression analysis indicated that the RPN classification correlated very well with the surgical difficulty score (R2 = 0.80). The R2 values for the other scoring systems were: 0.66 for RENAL, 0.75 for PADUA, and 0.70 for SPARE. In an external validation cohort, the performance of all four classification systems in predicting perioperative outcomes was similar, with low R2 values. Conclusions: The proposed RPN classification is the first nephrometry system to assess the surgical difficulty of renal masses for which robot-assisted PN is planned, and is a useful tool to assist in surgical planning, training and data reporting. Patient summary: We describe a simple classification system to help urologists in preoperative assessment of the difficulty of robotic surgery for partial kidney removal for kidney tumours.

Impact of Different Artificial Intelligence User Interfaces on Lung Nodule and Mass Detection on Chest Radiographs.

Purpose: To explore the impact of different user interfaces (UIs) for artificial intelligence (AI) outputs on radiologist performance and user preference in detecting lung nodules and masses on chest radiographs. Materials and Methods: A retrospective paired-reader study with a 4-week washout period was used to evaluate three different AI UIs compared with no AI output. Ten radiologists (eight radiology attending physicians and two trainees) evaluated 140 chest radiographs (81 with histologically confirmed nodules and 59 confirmed as normal with CT), with either no AI or one of three UI outputs: (a) text-only, (b) combined AI confidence score and text, or (c) combined text, AI confidence score, and image overlay. Areas under the receiver operating characteristic curve were calculated to compare radiologist diagnostic performance with each UI with their diagnostic performance without AI. Radiologists reported their UI preference. Results: The area under the receiver operating characteristic curve improved when radiologists used the text-only output compared with no AI (0.87 vs 0.82; P < .001). There was no difference in performance for the combined text and AI confidence score output compared with no AI (0.77 vs 0.82; P = .46) and for the combined text, AI confidence score, and image overlay output compared with no AI (0.80 vs 0.82; P = .66). Eight of the 10 radiologists (80%) preferred the combined text, AI confidence score, and image overlay output over the other two interfaces. Conclusion: Text-only UI output significantly improved radiologist performance compared with no AI in the detection of lung nodules and masses on chest radiographs, but user preference did not correspond with user performance.Keywords: Artificial Intelligence, Chest Radiograph, Conventional Radiography, Lung Nodule, Mass Detection© RSNA, 2023.

Evaluating the effectiveness of different perceptual training methods in a difficult visual discrimination task with ultrasound images.

Recent work has shown that perceptual training can be used to improve the performance of novices in real-world visual classification tasks with medical images, but it is unclear which perceptual training methods are the most effective, especially for difficult medical image discrimination tasks. We investigated several different perceptual training methods with medically naïve participants in a difficult radiology task: identifying the degree of hepatic steatosis (fatty infiltration of the liver) in liver ultrasound images. In Experiment 1a (N = 90), participants completed four sessions of standard perceptual training, and participants in Experiment 1b (N = 71) completed four sessions of comparison training. There was a significant post-training improvement for both types of training, although performance was better when the trained task aligned with the task participants were tested on. In both experiments, performance initially improves rapidly, with learning becoming more gradual after the first training session. In Experiment 2 (N = 200), we explored the hypothesis that performance could be improved by combining perceptual training with explicit annotated feedback presented in a stepwise fashion. Although participants improved in all training conditions, performance was similar regardless of whether participants were given annotations, or underwent training in a stepwise fashion, both, or neither. Overall, we found that perceptual training can rapidly improve performance on a difficult radiology task, albeit not to a comparable level as expert performance, and that similar levels of performance were achieved across the perceptual training paradigms we compared.

CLiP, catheter and line position dataset.

Correct catheter position is crucial to ensuring appropriate function of the catheter and avoid complications. This paper describes a dataset consisting of 50,612 image level and 17,999 manually labelled annotations from 30,083 chest radiographs from the publicly available NIH ChestXRay14 dataset with manually annotated and segmented endotracheal tubes (ETT), nasoenteric tubes (NET) and central venous catheters (CVCs).

Skin-Liver Distance and Interquartile Range-Median Ratio as Determinants of Interoperator Concordance in Acoustic Radiation Force Impulse Imaging.

CONTEXT AND AIMS: The accuracy of acoustic radiation force impulse (ARFI) ultrasound compared to liver biopsy is higher when there is concordance between F-scores of two or more operators. We hypothesized that when the first operator interquartile range/median-velocity ratio (IMR) is <0.3 and skin-liver distance (SLD) is <2.5 cm, there is greater interoperator concordance and a second operator is not necessary. SUBJECTS AND METHODS: Two-operator ARFI ultrasound measurements (F-score, SLD, and IMR) were recorded for 927 consecutive patients. Chi-squared testing compared interoperator concordance for SLD <2.5 cm versus SLD ≥2.5 cm and IMR <0.3 versus IMR ≥0.3 when SLD <2.5 cm, in each of the F-score groups of 0/1, 2, 3, and 4. RESULTS: Statistically significant differences were demonstrated between SLD <2.5 cm and SLD ≥2.5 cm groups for F-scores 0/1 or 4 (P = 0.005) and F-scores 2 or 3 (P < 0.001). Concordance, when SLD measured <2.5 cm, was more than 85% for all F-score groups. In the SLD <2.5 cm group, concordance fell below 85% when IMR ≥0.3, for all F-scores except F2. Specifically, P values comparing IMR <0.3 and IMR ≥0.3 in the various first operator F-score groups were P = 0.040 for F0/F1, P = 0.580 for F2, P = 0.342 for F3, and P < 0.001 for F4. CONCLUSIONS: ARFI measurements from one operator can be considered acceptable when SLD <2.5 cm and IMR <0.3. Otherwise, adding a second operator can improve confidence in the result.