Background/Objectives: The aim was to assess whether a machine learning (ML) algorithm could empower the ability of ultrasound (US) integrated with shear-wave elastography (SWE) to preoperatively define the ALN status in breast cancer (BC). Methods: Patients with at least one histologically proven BC lesion, who underwent preoperative breast US and SWE were retrospectively enrolled. BC lesions were segmented on US and SWE images by three different operators and radiomics features were extracted. A multi-step US and SWE feature selection was performed. A Simple Logistic ML classifier was applied to the dataset to predict the ALN status, its performance assessed through the AUC and Matthews Correlation Coefficient (MCC). The performance of the ML classifier was compared to that of an expert radiologist, who evaluated the US B-mode lymph-node features included in the test set. Results: A total of 133 BC lesions were included and divided into a training set, composed of 89 BC lesions (ALN-: 52; ALN+: 37), and a test set, including 44 BC lesions (ALN-: 24; ALN+: 20). Eight features out of the 1098 radiomics features extracted from US and SWE images were selected to build the predictive model. Simple Logistic classifier showed AUC of 0.685 and 0.677, MCC of 0.387 and 0.375 in the training and test set, respectively. The performance of the expert radiologist was higher than that of the ML classifier (AUC = 0.817), but not significantly different (p = 0.481). Conclusions: The inclusion of SWE-derived radiomics features could aid in the preoperative assessment of ALN status in BC using an ML approach.

This editorial offers commentary on the article which aimed to forecast the likelihood of short-term major postoperative complications (Clavien-Dindo grade ≥ III), including anastomotic fistula, intra-abdominal sepsis, bleeding, and intestinal obstruction within 30 days, as well as prolonged hospital stays following ileocecal resection in patients with Crohn’s disease (CD). This prediction relied on a machine learning (ML) model trained on a cohort that integrated a nomogram predictive model derived from logistic regression analysis and a random forest (RF) model. Both the nomogram and RF showed good performance, with the RF model demonstrating superior predictive ability. Key variables identified as potentially critical include a preoperative CD activity index ≥ 220, low preoperative serum albumin levels, and prolonged operation duration. Applying ML approaches to predict surgical recurrence have the potential to enhance patient risk stratification and facilitate the development of preoperative optimization strategies, ultimately aiming to improve post-surgical outcomes. However, there is still room for improvement, particularly by the inclusion of additional relevant clinical parameters, consideration of medical therapies, and potentially integrating molecular biomarkers in future research efforts.

PurposeTo evaluate the impact of a four-month training program on radiology residents' diagnostic accuracy in assessing deep myometrial invasion (DMI) in endometrial cancer (EC) using MRI.MethodThree radiology residents with limited EC MRI experience participated in the training program, which included conventional didactic sessions, case-centric workshops, and interactive classes. Utilizing a training dataset of 120 EC MRI scans, trainees independently assessed subsets of cases over five reading sessions. Each subset consisted of 30 scans, the first and the last with the same cases, for a total of 150 reads. Diagnostic accuracy metrics, assessment time (rounded to the nearest minute), and confidence levels (using a 5-point Likert scale) were recorded. The learning curve was obtained plotting the diagnostic accuracy of the three trainees and the average over the subsets. Anatomopathological results served as the reference standard for DMI presence.ResultsThe three trainees exhibited heterogeneous starting point, with a learning curve and a trend to more homogeneous performance with training. The diagnostic accuracy of the average trainee raised from 64 % (56 %-76 %) to 88 % (80 %-94 %) across the five subsets (p < 0.001). Reductions in assessment time (5.92 to 4.63 min, p < 0.018) and enhanced confidence levels (3.58 to 3.97, p = 0.12) were observed. Improvements in sensitivity, specificity, positive predictive value, and negative predictive value were noted, particularly for specificity which raised from 56 % (41 %-68 %) in the first to 86 % (74 %-94 %) in the fifth subset (p = 0.16). Although not reaching statistical significance, these advancements aligned the trainees with literature performance benchmarks.ConclusionsThe structured training program significantly enhanced radiology residents' diagnostic accuracy in assessing DMI for EC on MRI, emphasizing the effectiveness of active case-based training in refining oncologic imaging skills within radiology residency curricula.

Background: Differentiated thyroid cancer (DTC) patients have an outstanding overall long-term survival rate, and certain subsets of DTC patients have a very high likelihood of disease recurrence. Radioactive iodine (RAI) therapy is a cornerstone in DTC management, but cancer cells can eventually develop resistance to RAI. Radioactive iodine-refractory DTC (RAIR-DTC) is a condition defined by ATA 2015 guidelines when DTC cannot concentrate RAI ab initio or loses RAI uptake ability after the initial therapy. The RAIR condition implies that RAI cannot reveal new met-astatic foci, so RAIR-DTC metabolic imaging needs new tracers. 18F-FDG PET/CT has been widely used and has demonstrated prognostic value, but 18F-FDG DTC avidity may remain low. Fibroblast activation protein inhibitors (FA-Pi)s, prostatic-specific membrane antigen (PSMA), and somatostatin receptor (SSTR) tracers have been proposed as theragnostic agents in experimental settings and Arg-Gly-Asp (RGD) peptides in the diagnostic trial field. Multi-targeted tyrosine kinase inhibitors are relatively new drugs approved in RAIR-DTC therapy. Despite the promising targeted setting, they relate to frequent adverse-event onset. Sorafenib and trametinib have been included in re-differentiation protocols aimed at re-inducing RAI accumulation in DTC cells. Results appear promising, but not excellent. Conclusions: RAIR-DTC remains a challenging nosological entity. There are still controversies on RAIR-DTC definition and post-RAI therapy evaluation, with post-therapy whole-body scan (PT-WBS) the only validated criterion of response. The recent introduction of multiple diagnostic and therapeutic agents obliges physicians to pursue a multidisciplinary approach aiming to correct drug introduction and timing choice.

Purpose The aim of this retrospective study was to compare the MRI features between typical and atypical pheochromocytomas (Pheos) to specifically illustrate MRI features of atypical tumors for helping tumor diagnosis. Methods A total of 22 patients (14 women and 8 men, median age: 53 years, age range: 25–82 years) with Pheos evaluated using a 3 T MRI scanner were retrospectively collected; in particular, all patients had one tumor lesion, except in two cases who had two and three lesions, respectively, for a total of 25 tumor lesions. Results Of the total 25 tumor lesions included in our series, 12 lesions were classified as typical for their classical appearance on MRI (T1 hypointensity, T2 hyperintensity, no signal drop on T1 out-of-phase, restricted diffusion and persistent contrast enhancement). Conversely, the other 13 tumors were classified as having atypical lesions because they did not show the MRI features observed in typical Pheos; in particular, 3 lesions showed signal intensity suggestive of tumor hemorrhagic changes , 2 lesions were totally cystic with an internal fluid–fluid level and a thin capsula, 3 lesions showed predominantly cystic signal intensity with residual solid tissue in the peripheral capsula, and the remaining 5 lesions appeared as rounded partially cystic lesions with associated areas of solid tissue. Conclusion The imaging characterization of typical Pheos may be performed using MRI with specific imaging features; however, atypical Pheos represents a diagnostic challenge using MRI; in these tumors, cystic, necrotic, hemorrhagic, or fat changes may occur; thus, diagnostic pitfalls should be taken into consideration for MRI interpretation of such tumor type in clinical practice.

Objectives: To evaluate the effects on vascular enhancement of either a fixed rate (FR) or a fixed injection duration (FID) in single-pass (SP) contrast-enhanced abdominal multi-detector CT (CE-MDCT). Methods: Ninety-nine (54 M; 45 F; aged 18-86 years) patients with nontraumatic acute abdomen underwent a SP CE-MDCT after i.v. injection of 1.7 cc/Kg of a nonionic iodinated contrast media (370 mgI/mL) performed with either a FR (2 cc/s; Group A) or a FID (55 s; Group B). In both groups, patients were further stratified according to total body weight (kg) as follows: 40-60 (L); 61-80 (M); 81-100 (H). Signal- (SNR) and contrast-to-noise ratios (CNR) were calculated for the liver and for both abdominal aorta (AA) and main portal vein (MPV). Statistical analysis was performed by Student t- or Chi-square test for continuous and categorical data, respectively, whereas post hoc analysis was performed by the Mann-Whitney test (P < .05). Results: There were no significant differences in demographic and physical characteristics between Group A (n = 50; 53 +/- 20 years; BMI = 23.4 +/- 4.4) and Group B (n = 50; 51 +/- 17 years; BMI 22.7 +/- 4.2). Whereas overlapping findings were observed in the M sub-groups (n = 40), SNR and CNR were significantly higher (P < .01) in Group B for both AA and MPV in the high (H) weight sub-groups (n = 20) while not significant differences were observed in the low (L) weight sub-groups (n = 40) despite a significantly lower injection rate (1.6 +/- 0.2 cc/s, P < .01) in Group B. Conclusion: A FID results in an overall better vascular enhancement than a FR in SP CE-MDCT. Advances in knowledge: Single-pass is an optimized contrast-enhanced abdominal CT protocol combining the benefits of vascular and visceral enhancement and characterized by a customized scan delay tailored around a monophasic contrast injection. In single-pass protocol, a fixed injection duration (55 s) results in an overall better vascular enhancement than a fixed rate (2 cc/s) and should be therefore regarded as the injection modality of choice.

Background: Patients with differentiated thyroid cancer (DTC) are referred to radioactive 131I (RAI) therapy and post-therapy 131I whole-body scintigraphy (WBS) to identify local and/or remote metastases. Positron emission tomography (PET)/computed tomography (CT) imaging with 18F-fluoro-D-glucose (FDG) or 18F-sodium fluoride (NaF) may also be used with these patients for the evaluation of bone metastases. We compared the role of 18F-NaF PET/CT and 18F-FDG-PET/CT in patients with DTC and documented bone metastases at post-therapy WBS. Methods: Ten consecutive DTC patients with iodine avid bone metastasis at post-therapy WBS referred to 18F-NaF PET/CT and 18F-FDG PET/CT were studied. The findings of the three imaging procedures were compared for abnormal detection rates and concordance. Results: At post-therapy 131I WBS, all patients had skeletal involvement with a total of 21 bone iodine avid lesions. At 18F-FDG PET/TC, 19 bone lesions demonstrated increased tracer uptake and CT pathological alterations, while 2 lesions did not show any pathological finding. At 18F-NaF PET/CT, the 19 bone lesions detected at 18F-FDG PET/TC also demonstrated abnormal tracer uptake, and the other 2 bone iodine avid foci did not show any pathological finding. Conclusions: In patients with DTC, 18F-NaF PET/CT did not obtain more information on the metastatic skeletal involvement than post-therapy 131I WBS and 18F-FDG PET/CT.

PURPOSE:To build and validate a combined radiomics and machine learning (ML) approach using B-mode US and SWE images to differentiate benign from malignant solid breast lesions (BLs) and compare its performance with that of an expert radiologist. METHODS:Patients with at least one BI-RADS 2-6 BL who performed breast US integrated with SWE were retrospectively included. B-mode US and SWE images were manually segmented to extract radiomics features. A multi-step feature selection process was performed and a predictive model built using the Logistic Regression algorithm. The diagnostic accuracy was evaluated with the AUC and Matthews Correlation Coefficient (MCC) metrics. The performance of the ML classifier was compared to that of an expert radiologist. RESULTS:427 Bls were included and divided into a training (286 BLs, of which 127 benign and 159 malignant) and a test set (141 BLs, of which 59 benign and 82 malignant). Of 1098 features extracted from B-mode US and SWE images, 13 were finally selected. The ML classifier showed an AUC of 0.768 and 0.746, and an MCC of 0.403 and 0.423 in the training and test sets, respectively. The performance was higher than that of the expert radiologist assessing only B-mode US images, but significantly lower when SWE images were also provided. CONCLUSION:A ML approach based on B-mode US and SWE images may represent a potential tool in the characterization of BLs. SWE still gives its most relevant contribution in the clinical setting rather than included in a radiomics pipeline.

Second primary malignancies (SPM) are described as any primary, not synchronous, malignancy arising in a different anatomical district, with confirmed histological diagnosis. Age at diagnosis, previous non-thyroidal primary malignancy, and radioactive iodine (RAI) therapy have been proposed as independent risk factors for SPM. RAI therapy is a standard treatment for moderate-high risk differentiated thyroid cancer (DTC), and its effect on the development of SPM has become a critical topic in DTC treatment. The purpose of this retrospective single-center study was to investigate the occurrence and the possible association of non-thyroidal SPM diagnosed after DTC and RAI therapy in a cohort of 1326 consecutive DTC patients referred at our Institution for RAI treatment from 1993 to 2009. Eighty-nine patients with ages ≤ 18 years at the time of DTC diagnosis or with a follow-up of ≤12 months were excluded from the final analysis. All patients underwent a complete clinical and hematological follow-up every 6 months for a minimum of 12 months. During follow-up (mean 89 ± 73 months), 25 patients (2%) had an SPM diagnosis (mean 133 ± 73 months). The most common site of the second malignancy was the breast, accounting for 32% of all SPM, followed by colon-rectal cancer (16%), leukemia, and gynecological and kidney cancer (4%). At Cox univariable regression analysis, age at DTC diagnosis (p < 0.001), age ≥55 years (p < 0.001) and follow-up duration (p < 0.004) were associated with SPM onset, while no significant association was observed with the administered activity of radioiodine. In conclusion, our data suggest that the older a person gets, the more sharply the likelihood of developing additional diseases, such as PMS, increases. Similarly, for follow-up, the more a patient is followed up clinically over time, the higher the risk of new diagnoses increases.

PURPOSE:To build and validate a predictive model of placental accreta spectrum (PAS) in patients with placenta previa (PP) combining clinical risk factors (CRF) with US and MRI signs.METHOD:Our retrospective study included patients with PP from two institutions. All patients underwent US and MRI examinations for suspicion of PAS. CRF consisting of maternal age, cesarean section number, smoking and hypertension were retrieved. US and MRI signs suggestive of PAS were evaluated. Logistic regression analysis was performed to identify CRF and/or US and MRI signs associated with PAS considering histology as the reference standard. A nomogram was created using significant CRF and imaging signs at multivariate analysis, and its diagnostic accuracy was measured using the area under the binomial ROC curve (AUC), and the cut-off point was determined by Youden's J statistic.RESULTS:A total of 171 patients were enrolled from two institutions. Independent predictors of PAS included in the nomogram were: 1) smoking and number of previous CS among CRF; 2) loss of the retroplacental clear space at US; 3) intraplacental dark bands, focal interruption of the myometrial border and placental bulging at MRI. A PAS-prediction nomogram was built including these parameters and an optimal cut-off of 14.5 points was identified, showing the highest sensitivity (91%) and specificity (88%) with an AUC value of 0.95 (AUC of 0.80 in the external validation cohort).CONCLUSION:A nomogram-based model combining CRF with US and MRI signs might help to predict PAS in PP patients, with MRI contributing more than US as imaging evaluation.

Kasai portoenterostomy (KP) plays a crucial role in the treatment of biliary atresia (BA). The aim is to correlate MRI quantitative findings of native liver survivor BA patients after KP with a medical outcome. Thirty patients were classified as having ideal medical outcomes (Group 1; n = 11) if laboratory parameter values were in the normal range and there was no evidence of chronic liver disease complications; otherwise, they were classified as having nonideal medical outcomes (Group 2; n = 19). Liver and spleen volumes, portal vein diameter, liver mean, and maximum and minimum ADC values were measured; similarly, ADC and T2-weighted textural parameters were obtained using ROI analysis. The liver volume was significantly (p = 0.007) lower in Group 2 than in Group 1 (954.88 ± 218.31 cm3 vs. 1140.94 ± 134.62 cm3); conversely, the spleen volume was significantly (p < 0.001) higher (555.49 ± 263.92 cm3 vs. 231.83 ± 70.97 cm3). No differences were found in the portal vein diameter, liver ADC values, or ADC and T2-weighted textural parameters. In conclusion, significant quantitative morpho-volumetric liver and spleen abnormalities occurred in BA patients with nonideal medical outcomes after KP, but no significant microstructural liver abnormalities detectable by ADC values and ADC and T2-weighted textural parameters were found between the groups.

Abstract Purpose: Illustrate imaging findings of gastrinomas and non-functioning pancreatic endocrine tumors (NF-PNET) in a patient with multiple endocrine neoplasia type-1 (MEN-1) syndrome with a radiologic-pathologic correlation for both along with the results of a 13 yrs observational study. Methods: A 48 yrs old male patient with MEN-1 and a Zollinger-Ellison syndrome was submitted to a duodeno-cephalopancreatectomy (DCP) extended to the pancreatic body to remove several gastrinomas shown by an endoscopic-ultrasonography as well as a large (> 2 cm) hypo-vascular pancreatic nodule shown by a contrast-enhanced multi-detector CT (CE-MDCT). Further conventional (CT/MR) and functional imaging (68Ga-PET-DOTA-TOC) studies were performed over the next 13 years. Results: Up to 14 gastrin-positive NET-G1 (pT2,N1) as well as a single PNET-G2 (pT2,N0) were found at histo-pathology which also showed a NET-G1 in the uncinate process where CE-MDCT documented a 9 mm hyper-vascular nodule. A 7 mm pancreatic nodule with identical contrast-enhancement pattern was also shown at the level of the pancreatic tail which was left to preserve endocrine function. At this level, follow-up studies documented the occurence of a small (< 1 cm) hypo-vascular nodule which was metastatic at presentation and rapidly progressed under somastatin-analogs therapy whereas the hyper-vascular nodule remained stable over 13 years. Both the pancreatic lesion as well as the hepatic metastasis showed pathologic uptake of the radiotracer with a SUVmax of 6.3 and 29.5, respectively, allowing the patient to be scheduled for a Peptide Receptor Radionuclide Therapy performed with 29.6 GBq of 177Lu-Oxotreotide. Conclusions: Contrast-enhancement patterns are correlated with both the histological grade as well as the biological behaviour of PNETS.

BACKGROUND:Diagnosis of Crohn's disease (CD) requires ileo-colonoscopy (IC) and cross-sectional evaluation. Recently, "echoscopy" has been used effectively in several settings, although data about its use for CD diagnosis are still limited. Our aim was to evaluate the diagnostic accuracy of handheld bowel sonography (HHBS) in comparison with magnetic resonance enterography (MRE) for CD diagnosis.METHODS:From September 2019 to June 2021, we prospectively recruited consecutive subjects attending our third level IBD Unit for suspected CD. Patients underwent IC, HHBS, and MRE in random order with operators blinded about the result of the other procedures. Bivariate correlation between MRE and HHBS was calculated by Spearman coefficient (r). To test the consistency between MRE and HHBS for CD location and complications, the Cohen's k measure was applied.RESULTS:Crohn's disease diagnosis was made in 48 out of 85 subjects (56%). Sensitivity, specificity, positive predictive values, and negative predictive values for CD diagnosis were 87.50%, 91.89%, 93.33%, and 85% for HHBS; and 91.67%, 94.59%, 95.65%, and 89.74% for MRE, without significant differences in terms of diagnostic accuracy (89.41% for HHBS vs 92.94% for MRE, P = NS). Magnetic resonance enterography was superior to HHBS in defining CD extension (r = 0.67; P < .01) with a better diagnostic performance than HHBS for detecting location (k = 0.81; P < .01), strictures (k = 0.75; P < .01), abscesses (k = 0.68; P < .01), and fistulas (k = 0.65; P < .01).CONCLUSION:Handheld bowel sonography and MRE are 2 accurate and noninvasive procedures for diagnosis of CD, although MRE is more sensitive in defining extension, location, and complications. Handheld bowel sonography could be used as effective ambulatory (or out-of-office) screening tool for identifying patients to refer for MRE examination due to high probability of CD diagnosis.

The widespread use of cross-sectional imaging modalities, such as computed tomography (CT) and magnetic resonance imaging (MRI), in the evaluation of abdominal disorders has significantly increased the number of incidentally detected adrenal abnormalities, particularly adrenal masses [...].

Endometrial cancer (EC) is intricately linked to obesity and diabetes, which are widespread risk factors. Medical imaging, especially magnetic resonance imaging (MRI), plays a major role in EC assessment, particularly for disease staging. However, the diagnostic performance of MRI exhibits variability in the detection of clinically relevant prognostic factors (e.g., deep myometrial invasion and metastatic lymph nodes assessment). To address these challenges and enhance the value of MRI, radiomics and artificial intelligence (AI) algorithms emerge as promising tools with a potential to impact EC risk assessment, treatment planning, and prognosis prediction. These advanced post-processing techniques allow us to quantitatively analyse medical images, providing novel insights into cancer characteristics beyond conventional qualitative image evaluation. However, despite the growing interest and research efforts, the integration of radiomics and AI to EC management is still far from clinical practice and represents a possible perspective rather than an actual reality. This review focuses on the state of radiomics and AI in EC MRI, emphasizing risk stratification and prognostic factor prediction, aiming to illuminate potential advancements and address existing challenges in the field.