AIM: This study investigates the use of multiparametric magnetic resonance imaging (mp-MRI)-based radiomics for assessing perineural invasion (PNI) in rectal cancer. MATERIALS AND METHODS: A retrospective analysis was pe...AIM: This study investigates the use of multiparametric magnetic resonance imaging (mp-MRI)-based radiomics for assessing perineural invasion (PNI) in rectal cancer. MATERIALS AND METHODS: A retrospective analysis was performed on clinical and MRI data from 423 rectal cancer patients with confirmed surgical pathology, gathered from two centres. Of these, 343 patients from centre 1 were divided into a training set and an internal validation (in-vad) set in an 8:2 ratio, while 80 patients from centre 2 were used for independent external validation (ex-vad). Univariate and multivariate analyses were conducted on clinical features to build a clinical model. A combined model integrating both clinical and radiomic features was developed. RESULTS: Among all patients, 131 cases (31.0 %) were PNI-positive. A multivariate analysis revealed MRI-reported T (mrT) stage (odds ratio [OR] = 1.66, P=.010) and MRI-reported N (mrN) stage (OR = 1.91, P=.002) as independent predictors of PNI, forming the clinical model. After selecting radiomic features, 30 features were used to construct the radiomics model. The area under the curve (AUC) values for the clinical model in the training, in-vad, and ex-vad sets were 0.719, 0.631, and 0.760, respectively. The AUC values for the radiomics model in the training, in-vad, and ex-vad sets were 0.841, 0.815, and 0.916, respectively, while the AUC values for the combined model in the training, in-vad, and ex-vad sets showed AUC values of 0.899, 0.826, and 0.914, respectively. CONCLUSION: The mp-MRI-based radiomics model demonstrates high accuracy in predicting PNI status in rectal cancer, offering a noninvasive and reliable tool for preoperative assessment.
OBJECTIVE: To evaluate clinical performance and diagnostic accuracy of the MRI-based Osseous Tumour Reporting and Data System (OT-RADS) compared with traditional descriptive impressions. MATERIALS AND METHODS: After pros...OBJECTIVE: To evaluate clinical performance and diagnostic accuracy of the MRI-based Osseous Tumour Reporting and Data System (OT-RADS) compared with traditional descriptive impressions. MATERIALS AND METHODS: After prospective integration of OT-RADS into the electronic dictation system, fellowship-trained musculoskeletal radiologists applied OT-RADS scoring in routine bone tumour reporting. This retrospective review included consecutive patients from July 2021 to May 2024, aged 17-100 years, with extremity or body wall osseous lesions evaluated by MRI and confirmed by surgery, biopsy, or benign follow-up. Demographics, imaging findings, lesion characteristics, biopsy results, treatments, and outcomes were collected. Diagnostic accuracy was compared using McNemar's test. RESULTS: The study included 352 patients (60.2 % female, mean age 54.7 ± 15.7 years). Most lesions (66.1 %) were managed conservatively, while surgery and chemotherapy each accounted for 8.4 %. Biopsy was performed in 18.8 %. The distribution included OT-RADS I (4.8 %), II (46.1 %), III (18.2 %), and malignant-suspicion categories IV-V (36.1 %). Outcomes were favorable in 60.8 % of patients. OT-RADS demonstrated higher diagnostic accuracy than descriptive impressions (95.4 % vs. 88.0 %; p<.0001), particularly in biopsy-confirmed cases (84.6 % vs. 61.5 %; p<.0001). Accuracy was consistently superior across benign (II-III; p=.003) and malignant lesions (IV-V; p<.001), with subgroup analysis confirming significance for OT-RADS IV (p=.0078) and V (p=.0156). CONCLUSIONS: OT-RADS significantly outperforms descriptive reporting in accuracy for benign and malignant bone tumours. These results support broader adoption of OT-RADS as a standardized framework to stratify lesion risk, optimize management, and improve patient outcomes.
AIM: Diastolic dysfunction (DD) is central to hypertrophic cardiomyopathy (HCM), yet comprehensive cardiac magnetic resonance (CMR)-based staging is limited. We evaluated a multiparametric CMR framework combining time-vo...AIM: Diastolic dysfunction (DD) is central to hypertrophic cardiomyopathy (HCM), yet comprehensive cardiac magnetic resonance (CMR)-based staging is limited. We evaluated a multiparametric CMR framework combining time-volume indices, left atrial volume index (LAVI), and phase-contrast magnetic resonance imaging (PC-MRI) flow analysis against echocardiographic staging and healthy controls. MATERIALS AND METHODS: Fifty-four HCM patients and 65 matched controls underwent cine-CMR to derive peak filling rate (PFR), peak filling rate acceleration (PFRA), peak filling rate end time (PFRE), and peak ejection rate (PER). PC-MRI at the sinotubular junction yielded peak velocity, ejection time (ET), time to peak velocity (TPV), and TPV-to-ET ratio. LAVI was measured from 4-chamber cine. Serum N-terminal pro-B-type natriuretic peptide (NT-proBNP) and troponin I were analysed. DD was classified per American Society of Echocardiography / European Association of Cardiovascular Imaging (ASE/EACVI) early-to-late diastolic mitral inflow velocity ratio (E/A)-based criteria. Diagnostic performance was assessed by receiver operating characteristic (ROC) analysis; reproducibility was assessed by intraclass correlation coefficients (ICCs). RESULTS: DD was present in 70% of HCM patients. In controls, HCM showed a lower PFR (286.4 vs 397.8 ml/s), reduced PFRA (0.91 vs 1.48 ml/s), prolonged PFRE (321.3 vs 205.7 ms), and higher LAVI (48.3 vs 29.8 ml/m) (all P<0.001). Obstructive HCM demonstrated delayed systolic acceleration with a lower TPV-to-ET ratio (0.55 vs 0.67). LAVI and TPV-to-ET ratio each achieved an area under the curve (AUC) of 0.85 for DD discrimination, exceeding isolated time-volume indices. NT-proBNP correlated with DD severity and PFRE (r=0.42, P=0.01). PER decline indicated subclinical systolic impairment despite preserved ejection fraction (EF). Reproducibility was excellent (ICC: 0.92-0.95). CONCLUSION: A CMR framework integrating time-volume metrics, LAVI, and aortic flow kinetics enables accurate DD classification, detection of early systolic dysfunction, and characterisation of obstruction-related haemodynamics in HCM, supporting clinical application and longitudinal surveillance.
AIM: Exocrine pancreatic insufficiency (EPI) can be assessed noninvasively using magnetic resonance cholangiopancreatography (MRCP) with secretin stimulation. This pilot study measured pancreatic volumes in children with...AIM: Exocrine pancreatic insufficiency (EPI) can be assessed noninvasively using magnetic resonance cholangiopancreatography (MRCP) with secretin stimulation. This pilot study measured pancreatic volumes in children with suspected EPI and quantified total secreted fluid volume after secretin stimulation. MATERIALS AND METHODS: We retrospectively reviewed MRCPs performed between 2011 and 2021. Pancreatic volume was quantified using parametric magnetic resonance imaging (MRI) software on axial T2-weighted images. Secreted volume after secretin administration was measured with regions of interest on coronal single-shot fast spin echo (SSFSE) thick-slab sequences including thresholded pixels in the stomach and small intestine. RESULTS: We included 44 scans from 39 patients (23 males) in the study with a mean patient age of 10.7 ± 5.1 years. Median pancreas volume was 27.43 mL (interquartile range [IQR]: 15.92-53.51 mL). Pancreas volume correlated significantly with patient weight (Spearman's ρ = 0.63; P<0.001), height (ρ = 0.63; P<0.001), and body surface area (BSA) (ρ = 0.64; P<0.001). Median total secreted fluid volume after secretin was 71.35 mL (IQR: 31.77-119.56 mL). There was no statistically significant correlation of total secreted volume with patient age, weight, weight percentile, height percentile, body mass index (BMI), BMI percentile, or BSA (P>0.05). CONCLUSION: We report MRCP-derived pancreas volumes and total secreted fluid volumes in a cohort of children undergoing clinically indicated MRCP. Our study demonstrates the feasibility of obtaining such measures using MRI as a noninvasive tool. Pancreas volume correlated strongly with body size (weight, height, and BSA) and moderately with patient age, height percentile, and BMI. In contrast, total secreted volume showed no correlation with demographic or anthropometric measures.
AIM: The aim of our study was to assess pain during PRP injections and to evaluate correlations with pathology, anatomical location and needle size. The second goal was to evaluate the improvement in pain in the short an...AIM: The aim of our study was to assess pain during PRP injections and to evaluate correlations with pathology, anatomical location and needle size. The second goal was to evaluate the improvement in pain in the short and medium term. MATERIALS AND METHODS: Overall, 389 (195 males and 194 females) consecutive patients addressed for PRP injections were prospectively included. Pain during infiltration was evaluated using a visual analogue scale (VAS). Correlations with pathology, anatomical location and needle size were analyzed. Pain response was assessed in the short term, (10-15 days), and in the medium term (3 months). RESULTS: A total of 215 patients suffered from tendinopathy and 174 from osteoarthritis. Median age was 47.7 years old for tendinopathy and 46.03 years old for osteoarthritis. During injections, a significant difference was noticed between the groups (Mann-Whitney: p=0.0124). But no significant influence of the needle size or the anatomical location on recorded pain during injection (p=0.222). Improvement of short-term or medium-term pain values compared with the day of injection was not significant (p=0.3992, p=0.1202 respectively). CONCLUSION: Our study showed that PRP injections were associated with a higher level of pain in patients with tendinopathy than in those with arthritis, regardless of anatomical location or needle size. No improvement in pain was noticed in the short and medium term.
Radiomics allows for quantitative utilisation of radiological data and carries great potential for improving the diagnosis and management of hepatopancreatobiliary (HPB) cancers. Radiomic features represent changes at th...Radiomics allows for quantitative utilisation of radiological data and carries great potential for improving the diagnosis and management of hepatopancreatobiliary (HPB) cancers. Radiomic features represent changes at the mesoscopic scale and serve as noninvasive markers for tumour heterogeneity. Using deep learning and machine learning approaches, retrospective studies have demonstrated that radiomic signatures have the capability to improve the diagnosis of hepatocellular carcinoma (HCC), pancreatic cancer, and cholangiocarcinoma, in conjunction with radiological evaluation. Radiomic models have been successfully implemented to predict prognosis and treatment response, consistently outperforming established clinical markers. Novel pretreatment radiomic signatures predicting progression, survival and response to immunotherapy in advanced HCC demonstrate the great potential for radiomics in precision medicine. Correlation and integration of radiomics with genomic, metabolomic, and immunological data allows for noninvasive profiling of HPB cancers and the development of highly predictive integrated models. Future adoption of these works into clinical practice will allow for personalised diagnostic and treatment strategies. However, though these works show promise, further evaluation of optimal imaging strategies, image standardisation, and prospective validation across diverse patient populations is needed before widespread adoption in routine clinical practice.
AIM: The aim of this study was to systematically investigate protocols and performance of iopromide contrast-enhanced coronary computed tomography angiography (CCTA) for diagnosing coronary artery disease (CAD) in patien...AIM: The aim of this study was to systematically investigate protocols and performance of iopromide contrast-enhanced coronary computed tomography angiography (CCTA) for diagnosing coronary artery disease (CAD) in patients with atrial fibrillation (AF). MATERIALS AND METHODS: We searched PubMed, Embase, Web of Science, and the Cochrane Library for eligible studies of iopromide contrast-enhanced CCTA in patients with AF from inception to February 2025. Using a random-effect model, we pooled the following diagnostic metrics of CCTAs from included studies: diagnostic segments, effective radiation dose, sensitivity, specificity, and receiver operating characteristic (ROC) curve for CAD diagnosis versus conventional angiography. RESULTS: We included 14 studies involving 601 patients with AF who underwent iopromide 370 mgI/ml contrast-enhanced CCTA. Most studies applied individualised contrast and voltage protocols. Pooled analysis in 11 of those 14 studies indicated 98% segments were diagnostic (95% confidence interval [CI]: 0.96-0.99) with consistent results in subgroups of patients with heart rate variation >50 beats per minute, patients with overweight status, and patients with heart valve diseases. Scanners with ≥128 slices or individualised protocols could further improve image quality. The per-patient ROC was 0.97 (95% CI: 0.95-0.98), while the per-segment ROC was 0.99 (95% CI: 0.98-0.99). In addition, prospective-triggered or individualised voltage protocols reduced radiation dose by approximately 50%. CONCLUSION: Our meta-analysis showed that 370-mgI/ml iopromide contrast-enhanced CCTA delivers consistent image quality and accurate diagnosis of CAD in patients with AF. The use of high-performance scanners combined with individualised protocols and prospective acquisition could further enhance image quality while reducing radiation exposure.
AIM: The aim of this study was to assess the prevalence of a subcoracoid effusion and its association with rotator cuff tears at shoulder ultrasound. MATERIALS AND METHODS: This was a prospective study involving 3,405 pa...AIM: The aim of this study was to assess the prevalence of a subcoracoid effusion and its association with rotator cuff tears at shoulder ultrasound. MATERIALS AND METHODS: This was a prospective study involving 3,405 patients who underwent shoulder ultrasound referred from both primary and secondary care between January 2019 and January 2025. Demographic details, sonographic diagnosis including the size of a rotator cuff tear, and subcoracoid bursal distension was recorded. Details of further imaging studies and surgical outcomes were also documented. RESULTS: Of the 3405 patients examined, ultrasound revealed a subcoracoid effusion in 132 patients. Four patients had a subcoracoid effusion and no rotator cuff tear on ultrasound, and six patients reported a subcoracoid effusion and a partial thickness tear on ultrasound. A total of 122 (4%) patients had a subcoracoid effusion and a full-thickness tear on ultrasound, with 64 patients having surgical correlation (Fisher's exact test P<0.001). CONCLUSION: While subcoracoid effusions are not common, this study confirms a strong association between subcoracoid effusions and full-thickness rotator cuff tears. Careful assessment of the rotator cuff is essential when subcoracoid effusions are identified on ultrasound.
AIM: Massive epistaxis due to cavernous internal carotid artery (cICA) pathology is a life-threatening condition. The most common causes are craniofacial trauma and, rarely, spontaneous rupture of cICA aneurysms. While s...AIM: Massive epistaxis due to cavernous internal carotid artery (cICA) pathology is a life-threatening condition. The most common causes are craniofacial trauma and, rarely, spontaneous rupture of cICA aneurysms. While surgical management of cICA aneurysm is often challenging, endovascular methods offer a more straightforward and more reliable solutions. This article shares our institutional experience in the endovascular management of these grave pathologies and their long-term outcome. MATERIALS AND METHODS: We retrospectively identified 20 patients who presented with recurrent or massive epistaxis due to internal carotid artery pathologies (aneurysm/CCF). In our department, they were treated using endovascular methods. The outcome of the endovascular treatment was assessed based on the resolution of symptoms and the angiographic obliteration of ICA pathologies. RESULTS: Of the 20 patients in the study, 19 were male, and the mean age was 33.1 years (ranging from 19 to 72 years). The most common cause for epistaxis was craniofacial trauma due to road traffic accidents (RTA) (n = 14). Besides epistaxis, other clinical findings were ptosis (1), diplopia (1), tinnitus (7), proptosis (7), and carotid-cavernous fistula (CCF) (7). Treatment was instituted either as a reconstructive procedures (in 10 patients) or parent artery occlusion (PAO) (in 10 patients). CONCLUSION: Epistaxis, due to intracranial ICA pathologies is a life-threatening condition that needs urgent treatment. The plethora of causes underlying this aggressive condition can be safely mitigated using the endovascular route, which offers a more straightforward, reliable, and safer treatment compared with the surgical modality.
AIM: Current international guidance recommends semiannual ultrasound (US) for patients at high risk of developing hepatocellular carcinoma (HCC). This survey aims to understand current practices in the United Kingdom (UK...AIM: Current international guidance recommends semiannual ultrasound (US) for patients at high risk of developing hepatocellular carcinoma (HCC). This survey aims to understand current practices in the United Kingdom (UK) for HCC US surveillance and compare the findings with recently published standards for service delivery. MATERIALS AND METHODS: An online survey was advertised to UK US managers and department leads. Responses were collected between March and May in 2024. Survey questions were prioritised by the National Health Service (NHS) England HCC US surveillance working group and piloted in three departments. Eleven multiple choice and open questions were created to focus on HCC US surveillance staffing, training, protocols, and quality assurance. Findings were analysed using descriptive statistics to present frequencies and percentages. RESULTS: 43 centres from England, Scotland and Wales responded. Sonographers undertake HCC US surveillance in almost every centre (42/43; 97%) while half of respondents utilise radiologists to perform HCC US surveillance (23/32; 54%). Workforce capacity for HCC US surveillance was a concern for 35% (15/43) of respondents. There is wide variation in current scanning and reporting protocols; 53% (23/43) have no specific imaging protocol, 77% (33/43) have no reporting template. Most (32/43; 74%) centres had no designated HCC surveillance US lead radiologist or sonographer. Most did not provide specific surveillance training for US practitioners. Audit of HCC US surveillance quality and delivery was not undertaken in 60% (26/43) of responding centres. CONCLUSION: Our findings indicate a lack of consistency and governance across centres providing HCC US surveillance. Standardised training and guidance for individuals performing HCC US surveillance are required at a national level to ensure consistency in the quality of HCC US surveillance delivered across the UK.
AIM: This study aimed to investigate the relationship between the residual liver relative enhancement index (RL-REI), which is derived from gadoxetic acid (GA)-enhanced magnetic resonance imaging (MRI), and albumin-bilir...AIM: This study aimed to investigate the relationship between the residual liver relative enhancement index (RL-REI), which is derived from gadoxetic acid (GA)-enhanced magnetic resonance imaging (MRI), and albumin-bilirubin (ALBI) and Child-Pugh (CP) scores in assessing liver function in patients with hepatocellular carcinoma (HCC). MATERIALS AND METHODS: Patients with histopathologically confirmed HCC who underwent GA-enhanced MRI were retrospectively analysed. Receiver operating characteristic (ROC) curve analysis was conducted to identify the optimal cut-off values for the residual liver volume (RLV), the enhancement ratio (ER), and the RL-REI in predicting ALBI grades and CP classes. Spearman's rank correlation was used to assess the relationships of the RLV, the ER, and the RL-REI with the ALBI and CP scores. Intraclass correlation coefficient (ICC) was applied to evaluate intrareader reliability and inter-reader agreement. RESULTS: Forty-one patients (mean age: 56.7 ± 11.5 years, 33 men) met the study criteria. The RL-REI showed high diagnostic performance (area under the curve [AUC] ≥ 0.945 and accuracy ≥ 85.3%) for both readers in differentiating ALBI grades and CP classes (P<0.001). The RL-REI demonstrated a strong correlation with the ALBI and CP scores (r=-0.852 and -0.766 for reader 1, r=-0.839 and -0.758 for reader 2) (p<0.001). Intrareader and inter-reader agreement was almost perfect for all measurements (ICC ≥ 0.922) (P<0.001). CONCLUSION: The RL-REI demonstrated a strong correlation with ALBI and CP scores for assessing liver function in HCC patients and showed high accuracy in predicting ALBI grades and CP classes, highlighting its potential as a valuable radiological tool.
High-throughput extraction of quantitative image features, known as radiomics, has the potential to improve clinical radiology by revealing hidden features in medical images. In this review, we discuss the convergence of...High-throughput extraction of quantitative image features, known as radiomics, has the potential to improve clinical radiology by revealing hidden features in medical images. In this review, we discuss the convergence of radiomics with artificial intelligence (AI), specifically the development of large language models (LLMs) and agentic AI models, alongside improvements in radiomics methods, standardisation, and clinical uptake pathways. Our aim is to summarise these developments, highlight current challenges, and suggest potential future directions for the widespread adoption of radiomics in clinical practice. The study conducted an extensive literature review, focusing on radiomics research, particularly investigations that examined validation frameworks, standardisation efforts, deep learning, LLMs, multi-centre studies, and the emerging library of agentic pipelines. The review compares key publications to outline common findings affecting diagnostic accuracy, prognostic performance, reproducibility, and integration into clinical workflows. Advances in AI, particularly LLMs that work with both text and images to enhance understanding, and agentic AI systems that tailor workflows to specific conditions, offer promising opportunities to improve radiomic analysis and its clinical applicability. Issues of reproducibility are being addressed through standardisation and robust validation, paving the way for broader clinical implementation. Radiomics has the potential to become an essential component of precision imaging in the coming years, though this will require resolving challenges related to reproducibility, interpretability, and seamless integration into clinical practice through cross-disciplinary collaboration. CLINICAL RELEVANCE: Radiologists and imaging scientists should anticipate the future use of radiomic AI solutions, potentially incorporating LLMs, to support diagnostic, prognostic, and therapeutic decision-making, ultimately enhancing precision medicine.
AIM: This study aimed to evaluate the shear wave velocity (SWV) of supraspinatus tendons (SST) using Virtual Touch Imaging Quantification (VTIQ) elastography, to analyse factors influencing SWV measurements in healthy in...AIM: This study aimed to evaluate the shear wave velocity (SWV) of supraspinatus tendons (SST) using Virtual Touch Imaging Quantification (VTIQ) elastography, to analyse factors influencing SWV measurements in healthy individuals, and to characterise changes in SST stiffness during injury. MATERIALS AND METHODS: Clinical data from 130 healthy volunteers and 51 patients with SST injuries were analysed. The SWV values in healthy volunteers were assessed by VTIQ across gender, age, body mass index (BMI), and SST thickness groups. Patients were categorised by age (group 1: ≤50 years; group 2: >50 years). SST thickness and SWV values were measured bilaterally, with healthy volunteers serving as controls. Two-dimensional (2D) ultrasound acoustic characteristics were evaluated for injured SSTs. RESULTS: Bilateral SST SWV values showed no significant differences between sexes or SST thickness groups, but females had lower SWV values than males (P = 0.003). SWV values decreased with increasing age and BMI, with significant differences between age (P < 0.001) and BMI groups (P = 0.04). In individuals over 50 years old, right SST SWV values were lower than the left (P < 0.05). Injured SST in group 1 exhibited reduced echogenicity and significantly lower SWV values compared to controls (P < 0.05). Group 2 showed increased echogenicity with no significant difference in SWV values (P > 0.05). CONCLUSION: VTIQ elastography effectively reflects SST stiffness changes. Age and BMI significantly influence SWV values, while SST injuries result in increased thickness and reduced stiffness mainly. Overall, VTIQ elastography serves as a valuable complement to conventional 2D ultrasound for diagnosing SST injuries.
AIM: Radiation is the standard therapy for prostate cancer. However, it may cause damage to the surrounding organs, including the rectum. SpaceOAR hydrogel is a Food and Drug Administration (FDA)-approved spacer which in...AIM: Radiation is the standard therapy for prostate cancer. However, it may cause damage to the surrounding organs, including the rectum. SpaceOAR hydrogel is a Food and Drug Administration (FDA)-approved spacer which increases the distance between the rectum and prostate, reducing rectal radiation exposure. A 7.5 mm distance is considered desirable for radiation toxicity purposes. This study aims to quantify the increased recto-prostatic distance after SpaceOAR hydrogel placement. MATERIALS AND METHODS: A total of 118 consecutive patients who underwent magnetic resonance imaging (MRI) prior to and after SpaceOAR hydrogel placement were included. The distance between the prostate and rectum distance (PRD) was measured in nine different locations, including one centimetre superior to the apex, inferior to the base, and at the level of mid-gland on sagittal views, and one centimetre to the left, right, and midline of the prostate on axial views. The measured data were recorded in Microsoft Excel and compared before and after SpaceOAR hydrogel placement. RESULTS: The average PRD was significantly different before and after SpaceOAR hydrogel placement at all locations (P < 0.001). The mean PRD increased throughout the entire prostate (ranging from 8 ± 3 mm at base to 12 ± 14 mm at the apex). Although the averages were significantly different, up to 57% of patients did not achieve a PRD of 7.5 mm after SpaceOAR hydrogel placement. CONCLUSION: Although SpaceOAR hydrogel increases the mean PRD, the distribution of SpaceOAR hydrogel between the rectum and prostate may not be uniform, which may leave some portions of the rectum at increased risk for radiation toxicity.
AIM: This meta-analysis evaluates the diagnostic accuracy of computed tomography (CT)-based artificial intelligence (AI) in detecting lymph node metastasis in rectal cancer patients. MATERIALS AND METHODS: A thorough lit...AIM: This meta-analysis evaluates the diagnostic accuracy of computed tomography (CT)-based artificial intelligence (AI) in detecting lymph node metastasis in rectal cancer patients. MATERIALS AND METHODS: A thorough literature search was conducted in PubMed, Embase, and Web of Science to identify studies on the diagnostic accuracy of CT-based AI in detecting lymph node metastasis in rectal cancer patients, with a deadline of September 2024. The analysis assessed the performance of AI algorithms in detection through CT imaging. Pooled sensitivity and specificity were calculated using a bivariate random-effects model, reported with 95% confidence intervals. Study heterogeneity was evaluated using the I statistic to account for heterogeneity among the included studies. RESULTS: Seven studies were included in this meta-analysis. The internal test set contained 23 contingency tables, while the radiologists' data included 19 contingency tables. AI's pooled sensitivity, specificity, and area under the curve (AUC) were 0.84 (0.79-0.87), 0.77 (0.66-0.85), and 0.88 (0.16-1.00), respectively. Radiologists had pooled sensitivity, specificity, and AUC of 0.80 (0.72-0.86), 0.54 (0.46-0.63), and 0.72 (0.11-0.98). Funnel plot analysis showed P values of 0.13 for AI and 0.40 for radiologists, indicating that no publication bias was observed. CONCLUSION: CT-based AI demonstrates superior specificity over radiologists for LNM detection in rectal cancer. However, significant heterogeneity across studies and limited external tests of existing models warrant further multicenter test studies before clinical implementation.
AIM: United Kingdom medicolegal claims are rising. Many radiological allegations cite failure to report incidental findings, but these are often inconspicuous and asymptomatic and therefore difficult to detect. Neverthel...AIM: United Kingdom medicolegal claims are rising. Many radiological allegations cite failure to report incidental findings, but these are often inconspicuous and asymptomatic and therefore difficult to detect. Nevertheless, claimants often support their allegations by citing the Royal College of Radiologists standards for the interpretation and reporting of imaging investigations. We examined our medicolegal practice to determine the proportion where allegations cited incidental pathology. METHODS: Five medicolegal expert witnesses, working in abdominopelvic radiology, reviewed their medicolegal practice and determined the proportion of recent cases in which allegations of substandard reporting cited failure to diagnose incidental pathology. RESULTS: We identified 112 individual cases, comprising 176 imaging examinations; 99 (56 %) were computed tomography and 36 (20 %) magnetic resonance imaging. One hundred forty-four (82 %) were reported in the National Health Service, 14 (8 %) outsourced, and 18 (10 %) private practice. Medicolegal allegations cited failure to report incidental pathology in 41 (37 %) of the 112 patients. The commonest allegation in this group was failure to identify tumour (23 of 41 patients, 56 %). Interpretation was judged as substandard in 12 (29 %) of these patients compared with 23 (32 %) of the 71 patients in whom allegations did not cite incidental findings. CONCLUSION: A substantial proportion of medicolegal allegations in abdominal radiology cite failure to report incidental pathology. College standards should recognise that unanticipated pathology is sometimes difficult to detect, especially without a clear steer from the referring clinician.
AIM: We evaluated the performance of radiomics models based on multiparametric magnetic resonance imaging (MRI; dynamic contrast-enhanced, T2, and apparent diffusion coefficient (ADC) sequences) in differentiating malign...AIM: We evaluated the performance of radiomics models based on multiparametric magnetic resonance imaging (MRI; dynamic contrast-enhanced, T2, and apparent diffusion coefficient (ADC) sequences) in differentiating malignant from benign masses. Various combinations of MRI sequences were compared. MATERIALS AND METHODS: One hundred forty-four patients with breast lesions (69 malignant and 75 benign) were retrospectively analysed. After manual segmentation, 107 shape- and texture-based features were extracted from each lesion. Support vector classifier, random forest, and k-nearest neighbours were trained to distinguish malignant and benign lesions. The models were evaluated with seven different input image sets: five single-sequence images and two combined sets (all sequences vs all contrast-free sequences). RESULTS: Single-sequence models achieved similar accuracies for ADC, pre-contrast T1, and post-contrast T1, while T2 models showed slightly lower accuracy. The all-sequence model achieved an average accuracy of 0.88, outperforming the single-sequence models. The contrast-free sequence models (T2, ADC, and pre-contrast T1) yielded an average accuracy of 0.90, comparable to the all-sequence model. CONCLUSION: Non-contrast MRI radiomics algorithms have the potential for clinical implementation as complementary tools, potentially eliminating the need for contrast in MRI studies.
Breast lesions constitute a significant proportion of women's health imaging. While mammography, ultrasound, and magnetic resonance imaging (MRI) are well established methods of breast assessment, the radiation penalty a...Breast lesions constitute a significant proportion of women's health imaging. While mammography, ultrasound, and magnetic resonance imaging (MRI) are well established methods of breast assessment, the radiation penalty associated with computed tomography (CT) limits its role mainly to incidental detection of breast lesions, oncological staging, and follow-up. Spectral CT, including dual-energy CT and photon-counting CT enables advanced postprocessing techniques such as iodine maps, monoenergetic reconstructions, and material decomposition which may aid in lesion breast detection, characterisation, and oncological treatment response monitoring. This article illustrates the utility of spectral CT in a range of benign and malignant breast lesions, with correlation to conventional breast imaging. A series of benign and malignant breast lesions, alongside relevant iodine maps, monoenergetic reconstructions, and material decomposition will be demonstrated, with corresponding conventional breast imaging. The clinical applications in lesion detection, differentiation between benign and malignant lesions, staging, prognostication, and treatment response will be discussed. Whilst spectral CT will not replace conventional breast imaging methods, it has a role as a valuable adjunct to aid the radiologist in routine clinical practice. Emerging functional and quantitative capabilities have the potential for objective lesion assessment and further integration of CT into breast imaging practice.