In this paper, we present and discuss a novel reliability metric to quantify the extent a ground truth, generated in multi-rater settings, as a reliable basis for the training and validation of machine learning predictive models. To define this metric, three dimensions are taken into account: agreement (that is, how much a group of raters mutually agree on a single case); confidence (that is, how much a rater is certain of each rating expressed); and competence (that is, how accurate a rater is). Therefore, this metric produces a reliability score weighted for the raters’ confidence and competence, but it only requires the former information to be actually collected, as the latter can be obtained by the ratings themselves, if no further information is available. We found that our proposal was both more conservative and robust to known paradoxes than other existing agreement measures, by virtue of a more articulated notion of the agreement due to chance, which was based on an empirical estimation of the reliability of the single raters involved. We discuss the above metric within a realistic annotation task that involved 13 expert radiologists in labeling the MRNet dataset. We also provide a nomogram by which to assess the actual accuracy of a classification model, given the reliability of its ground truth. In this respect, we also make the point that theoretical estimates of model performance are consistently overestimated if ground truth reliability is not properly taken into account.

Musculoskeletal ultrasound involves the study of many superficial targets, especially in the hands, wrists, and feet. Many of these areas are within the first 3 cm of the skin surface and are ideal targets for ultra-high-frequency ultrasound. The high spatial resolution and the superb image quality achievable allow foreseeing a wider use of this novel technique, which has the potential to bring innovation to diagnostic imaging.

Small renal masses (SRM) are increasingly detected incidentally during imaging. They vary widely in histology and aggressiveness, and include benign renal tumors and renal cell carcinomas that can be either indolent or aggressive. Imaging plays a key role in the characterization of these small renal masses. While a confident diagnosis can be made in many cases, some renal masses are indeterminate at imaging and can present as diagnostic dilemmas for both the radiologists and the referring clinicians. This review focuses on CT characterization of small renal masses, perhaps helping us understand small renal masses. The following aspects were considered for the review: (a) assessing the presence of fat, (b) assessing the enhancement, (c) differentiating renal tumor subtype, and (d) identifying valuable CT signs.

Several studies have shown that ultrasound guidance may contribute to improved safety, effectiveness and accuracy of musculoskeletal interventional procedures performed around the hip if compared to those performed with a landmark-guided technique. Different approaches and injectates can be used for treating hip musculoskeletal disorders. These procedures may involve injections in the hip joint, periarticular bursae, tendons, and peripheral nerves. Intra-articular hip injections are mostly used as a conservative approach for treating patients affected by hip osteoarthritis. Ultrasound-guided injection of the iliopsoas bursa is performed in patients with bursitis and/or tendinopathy, to treat those with painful prosthesis due to iliopsoas impingement, or when the lidocaine test is indicated to identify the iliopsoas as a source of pain. Ultrasound-guided interventions are routinely used in patients with greater trochanteric pain syndrome having as target the gluteus medius/minimus tendons and/or the trochanteric bursae. Ultrasound-guided fenestration and platelet-rich plasma injection are applied in patients with hamstring tendinopathy with good clinical outcomes. Last but not least, ultrasound-guided perineural injections can be used for peripheral neuropathies or blocks of the sciatic, lateral femoral cutaneous, and pudendal nerves. In this paper, we discuss the evidence and technical tips for musculoskeletal interventional procedures performer around the hip, highlighting the added value of ultrasound as an imaging guidance modality.

The correct management of acute, subacute and overuse-related elbow pathologies represents a challenging diagnostic and therapeutic problem. While major trauma frequently requires a rapid surgical intervention, subluxation and minor trauma allow taking more time for diagnostics and planning the correct elective treatment after careful clinical and radiological investigation. In these conditions, communication between orthopaedic surgeon and radiologist allow to create a detailed radiology report, tailored to the patient's and surgeon's needs and optimal to plan proper management. Imaging technique as X-Ray, CT, US, MRI, CTA and MRA all belong to the radiologist's portfolio in elbow diagnostics. Detailed knowledge of elbow pathology and its classification and of the possibilities and limits of each imaging technique is of crucial importance to reach the correct diagnosis efficiently. The aim of this review is to present the most frequent elbow pathologies and suggest a suitable diagnostic approach for each of them.