Automatic segmentation of femoral tumors by nnU-net.

BACKGROUND: Metastatic femoral tumors may lead to pathological fractures during daily activities. A CT-based finite element analysis of a patient's femurs was shown to assist orthopedic surgeons in making informed decisions about the risk of fracture and the need for a prophylactic fixation. Improving the accuracy of such analyses ruqires an automatic and accurate segmentation of the tumors and their automatic inclusion in the finite element model. We present herein a deep learning algorithm (nnU-Net) to automatically segment lytic tumors within the femur. METHOD: A dataset consisting of fifty CT scans of patients with manually annotated femoral tumors was created. Forty of them, chosen randomly, were used for training the nnU-Net, while the remaining ten CT scans were used for testing. The deep learning model's performance was compared to two experienced radiologists. FINDINGS: The proposed algorithm outperformed the current state-of-the-art solutions, achieving dice similarity scores of 0.67 and 0.68 on the test data when compared to two experienced radiologists, while the dice similarity score for inter-individual variability between the radiologists was 0.73. INTERPRETATION: The automatic algorithm may segment lytic femoral tumors in CT scans as accurately as experienced radiologists with similar dice similarity scores. The influence of the realistic tumors inclusion in an autonomous finite element algorithm is presented in (Rachmil et al., "The Influence of Femoral Lytic Tumors Segmentation on Autonomous Finite Element Analyses", Clinical Biomechanics, 112, paper 106192, (2024)).

Ultrasound, magnetic resonance imaging and radiography of the finger joints in psoriatic arthritis patients.

OBJECTIVES: To report the discrepancies and agreements between US, MRI and radiography of the hand in PsA, and to compare the sensitivity and specificity of US and radiography to MRI as the gold standard imaging study in PsA. METHODS: All of the 100 prospectively recruited consecutive PsA patients underwent clinical assessment and concomitant radiographic, US and MRI studies of the MCP, PIP and DIP joints of one hand. Synovitis, flexor tenosynovitis, extensor paratenonitis, erosions and bone proliferations were identified and scored. All readers were blinded to clinical data, and agreement was calculated based on prevalence-adjusted bias-adjusted kappa (PABAK). RESULTS: The prevalence of synovitis, flexor tenosynovitis, extensor paratenonitis and erosions was similar for US and MRI, while that of bone proliferation was significantly increased in US and radiography compared with MRI (P < 0.001). The absolute agreement between US and MRI was good-to-very good for synovitis (85-96%, PABAK = 0.70-0.92), flexor tenosynovitis (93-98%, PABAK = 0.87-0.96) and extensor paratenonitis (95-98%, PABAK = 0.90-0.97). Agreement between US, MRI and radiography was 96-98% (PABAK = 0.92-0.97) for erosions and 71-93% (PABAK = 0.47-0.87) for bone proliferations. Sensitivity of US with MRI as gold standard was higher for synovitis (0.5-0.86) and extensor paratenonitis (0.63-0.85) than for flexor tenosynovitis (0.1-0.75), while the specificity was high for each pathology (0.89-0.98). CONCLUSION: There is very good agreement between US and MRI for the detection of inflammatory changes in finger joints in PsA. US, radiography and MRI have a good-to-very good agreement for destructive changes.

Left Atrial Volume as a Biomarker of Target Organ Damage in Cardionephrology: A Study in a Wide Range of Renal Function.

BACKGROUND: Cardiac chamber size was previously studied by echocardiography, among patients with progressive kidney disease only. We aimed to explore the relations between all heart chamber volumes as assessed by CT pulmonary angiography and renal function, throughout all ranges. METHODS: Using a novel technology for automatic four-chamber volumetric analysis, we analyzed CT pulmonary angiography results for 748 consecutive patients (between January 1 and December 31, 2014) without pulmonary embolism, and correlated the chamber volumes, indexed to body surface area, with the estimated glomerular filtration rates (eGFRs). RESULTS: After adjustment for age, all cardiac chamber volumes aside from that of the left ventricle inversely correlated with the eGFR. For every 10-unit decrease in eGFR, the volume indices of the left atrium, right atrium, and right ventricle were larger by 3.1%, 2.3%, and 1.5% (P < .001, P = .003, P = .007), respectively. There was a significant correlation between left atrial volume index and eGFR (r = -0.43; P < .001). In a categorical comparison, patients with even minimal renal dysfunction (eGFR, 60 to 90 mL/min/1.73 m2) had a 10.3% larger left atrial volume index (95% CI, 3.1-17.9; P = .004) compared to patients with an eGFR > 90 mL/min/1.73 m2. CONCLUSIONS: A simple concomitant volumetric analysis of all four cardiac chambers by CT pulmonary angiography demonstrated that differences in volume correlate with renal function even within the normal range spectrum. The difference was most evident in the left atrium. This finding may be the first clue to evolving cardiorenal syndrome and may serve as a target for early therapeutic interventions. TRIAL REGISTRY: ClinicalTrials.gov; Clinical Trials registration number (Helsinki Committee): 0603-15-TLV; URL: www.clinicaltrials.gov.