British Journal of Radiology最新文献

Radiomics is a health technology that has the potential to extract clinically meaningful biomarkers from standard of care imaging. Despite a wealth of exploratory analysis performed on scans acquired from patients with lung cancer and existing guidelines describing some of the key steps, no radiomic-based biomarker has been widely accepted. This is primarily due to limitations with methodology, data analysis and interpretation of the available studies. There is currently a lack of guidance relating to the entire radiomic workflow from study design to critical appraisal. This guide, written with early career lung cancer researchers, describes a more complete radiomic workflow. Lung cancer image analysis is the focus due to some of the unique challenges encountered such as patient movement from breathing. The guide will focus on CT imaging as these are the most common scans performed on patients with lung cancer. The aim of this article is to support the production of high-quality research that has the potential to positively impact outcome of patients with lung cancer.

Objective: To explore the value of native T1 and apparent diffusion coefficients (ADC) for predicting subtypes and stages of thymic epithelial tumors (TETs).

Methods: Sixty-seven patients with TETs confirmed by pathological analysis were retrospectively enrolled. The mean native T1 (T1mean), global native T1 (ADCtotal), relative minimum ADC (ADCmin), and global ADC (ADCtotal) values of the tumor were measured and compared for differences among low-risk thymoma (LRT), high-risk thymoma (HRT), and thymic carcinoma (TC). The differentiating efficacy was determined using receiver operating characteristic curve analysis.

Results: The native T1 values in thymoma or early stage were significantly higher than those in TC (P < 0.05/3) or advanced stage of TETs (P < 0.01). The ADC values in LRT or early stage were significantly higher than those in TC (P < 0.05/3) or advanced stage of TETs (P < 0.001). For differentiating the thymoma or HRT from TC, a combination of native T1 and ADC achieved the highest efficacy with an AUC of 0.891 and 0.851, respectively. For determining the TET stage, ADCmin achieved a relatively high diagnostic efficacy with an AUC of 0.933, and a combination of native T1 and ADC obtained an AUC of 0.876.

Conclusion: The combination of native T1 and ADC values could be helpful in clinical practice regarding evaluating TETs before treatment.

Advances in knowledge: The combination of T1 mapping and DWI can improve diagnostic accuracy and help guide clinical practice in providing the best individual treatment for TET patients.

Objectives: The splenial angle (SA), measured on axial DTI colour fractional anisotropy MRI, outperformed the callosal angle (CA) in predicting idiopathic normal pressure hydrocephalus (NPH) patients from those with Alzheimer's dementia, Parkinson's disease (PD) and healthy controls (HC). We investigated its reliability and classification performance on more commonly acquired T1 MPRAGE and FLAIR images.

Methods: SA was measured on axial MPRAGE and FLAIR images in 57 subjects (19 NPH, PD and HC each) by two raters, and compared across groups. Receiver operating characteristics (ROC) analysis was used to assess its classification performance differentiating NPH from non-NPH groups, in comparison to the CA.

Results: Inter-rater reliability for SA were excellent (intraclass correlation coefficients ≥0.91). SA was effective in differentiating NPH from non-NPH patients on MPRAGE and FLAIR images (p < 0.001). Its ROC curves showed excellent performance classifying NPH from HC (AUC 1) and PD (AUC >0.93) groups, and were highly comparable to those for CA (1; 0.947). Angles wider than 60° and narrower than 45° robustly (100%) excluded and predicted NPH from HC, respectively. The narrower 45° cutoff yielded better sensitivity (84.2-89.5%) in differentiating NPH from PD patients.

Conclusions: The SA on MPRAGE/FLAIR images showed excellent inter-rater reliability and classification performance predicting NPH from non-NPH groups, rivalling those of the CA.

Advances in knowledge: The SA on MPRAGE and FLAIR images is reproducible and shows excellent diagnostic performance differentiating NPH from non-NPH groups, with potential to replace the CA in NPH screening given its accessibility on routine axial neuroimaging.

CNS vascular syndromes are being increasingly recognized and diagnosed in the pediatric population. Their rarity and complexity makes diagnosis challenging. These syndromes can present in a number of different ways clinically and radiologically. A systematic method of image interpretation often reveals a pattern, allowing the diagnosis to be narrowed or further testing to be directed. Although the CNS vascular appearance itself is rarely specific, additional information gleaned from imaging other systems aids in pattern recognition. We retrospectively reviewed pediatric cases of CNS vasculopathy with a confirmed diagnosis of a syndromic disorder. We recorded the predominant CNS vascular appearance in each case, and assessed imaging clues from multiple systems, including brain, orbits, skull, spine, long bones, viscera and skin. Several of these syndromes also had systemic vascular involvement, which allowed further categorization. Some imaging patterns allowed a specific diagnosis to be made. In other cases, the constellations of imaging appearances helped distinguish between clinically similar phenotypes. The imaging pattern within each pediatric CNS vascular syndrome is fairly consistent, and awareness of these syndromes and expected imaging patterns allow the radiologist to suggest further imaging or targeted genetic testing.

Objectives: Radiation-induced lymphocyte apoptosis (RILA) and chromosomal damage assays (CDA) assays are proposed predictors of radiotherapy (RT) adverse events (RTAE). This study evaluated RILA and CDA in patients undergoing different RT dose regimens for early breast (FAST trial) or prostate (CHHiP trial) cancer.

Methods: Consecutive patients were recruited from each trial. Fresh heparinised blood samples were analyzed for RILA and CDA. The primary endpoint was time to first change in photographic breast appearance (FAST) or time to first grade ≥2 RTOG bladder or bowel toxicity (CHHiP). The secondary endpoint in FAST was breast fibrosis.

Results: The dataset included 103 FAST and 297 CHHiP trial patients. No significant association of RILA with the primary endpoint was observed in the FAST trial. However, the risk of grade ≥2 breast fibrosis was lower in patients with RILA ≥24% compared to those with RILA ≤16% (p = 0.012). In the CHHiP trial, no significant associations were found between CDA after prostate radiotherapy outcomes. However, higher levels of micronuclei per cell were associated with a lower risk of grade ≥2 RTOG pelvic toxicities. The relative risk of developing grade ≥2 RTAE decreased for patients with RILA ≥ 24% but was not statistically significant.

Conclusions: No association was found between RILA and photographic breast appearance. High RILA values were statistically associated with a lower risk of grade ≥2 breast fibrosis. In the CHHiP trial, most assays showed no association with pelvic toxicities.

Advances in knowledge: RILA is confirmed as a potential predictor of breast fibrosis regarding fraction sizes.

Vertebral fragility fractures (VFFs) are the most common osteoporotic fracture, with significant associated morbidity, mortality and risk of future fractures. Unfortunately, VFFs are frequently underdiagnosed and underreported. With this background, The Royal College of Radiologists (RCR), in collaboration with the Royal College of Physicians (RCP) and Royal Osteoporosis Society (ROS), undertook a UK-wide retrospective audit in 2019 to assess radiology reporting of incidental VFFs present on CT studies. The audit standards were not achieved; however, following a series of interventions, a re-audit in 2022 demonstrated generalised improvement in compliance with the audit parameters, in particular a 6-percentage point increase in moderate/severe VFFs reported. This article examines the potential benefits in terms of reductions in patient morbidity, mortality and costs that could be expected if the improvements identified were translated to a national level, acknowledging and discussing other improvements that need to occur, with particular reference to onward referral pathways.

Objectives: To develop a deep learning (DL) model based on ultrasound (US) images of lymph nodes for predicting cervical lymph node metastasis (CLNM) in postoperative patients with differentiated thyroid carcinoma (DTC).

Methods: Retrospective collection of 352 lymph nodes from 330 patients with cytopathology findings between June 2021 and December 2023 at our institution. The database was randomly divided into the training and test cohort at an 8:2 ratio. The DL basic model of longitudinal and cross-sectional of lymph nodes was constructed based on ResNet50 respectively, and the results of the two basic models were fused (1:1) to construct a longitudinal + cross-sectional DL model. Univariate and multivariate analysis were used to assess US features and construct a conventional US model. Subsequently, a combined model was constructed by integrating DL and US.

Results: The diagnostic accuracy of the longitudinal + cross-sectional DL model was higher than that of longitudinal or cross-sectional alone. The AUC of the combined model (US＋DL) was 0.855 (95%CI: 0.767-0.942), and the accuracy, sensitivity and specificity were 0.786 (95%CI: 0.671-0.875), 0.972 (95%CI: 0.855-0.999) and 0.588 (95%CI: 0.407-0.754), respectively. Compared with US and DL models, the IDI and NRI of the combined model are both positive.

Conclusions: This study preliminary shows that the DL model based on US images of lymph nodes has a high diagnostic efficacy for predicting CLNM in postoperative patients with DTC, and the combined model of US+DL is superior to single conventional US and DL for predicting CLNM in this population.

Advances in knowledge: We innovatively used DL of lymph node US images to predict the status of cervical lymph nodes in postoperative patients with DTC.

Objectives: To evaluate current UK practice for inflammatory bowel disease (IBD) imaging against recommendations from published international literature.

Methods: A retrospective multi-centre audit was undertaken evaluating imaging modalities, protocols and pathways used to investigate IBD both in outpatient and inpatient settings during January-December 2022. Reporting practices and training provisions were also recorded.

Results: Forty-one centres contributed: 35 centres provided complete data, 6 incomplete. Magnetic resonance enterography (MRE) was the most common modality for small bowel imaging across UK centres, comprising 13,099/18,784 (69.7%) investigations. There was regional variability in other modalities used, with 5 centres performing 81% of all intestinal ultrasound, and 3 centres performing 65% of all small bowel follow-through. Compared to outpatients, inpatients with suspected IBD were significantly more likely to be imaged with techniques imparting ionising radiation whether scanned either in-hours (p = 0.005) or out-of-hours (p < 0.001). Non-ionising radiation imaging modalities were significantly less available out-of-hours (p < 0.0001). Sequences included in MRE protocols were variable. Disparity in imaging follow-up for patients prescribed biologic therapies was observed.

Conclusions: Considerable variation in UK IBD imaging practice has been identified. Improvements must be made to reduce the regional inequality of patient access to different imaging modalities and decrease reliance on ionising radiation for inpatients. Further research to standardise and optimise imaging pathways should be undertaken to improve uniformity, with emphasis placed on training and education.

Advances in knowledge: This multi-centre audit showed considerable IBD imaging practice variation between UK centres, particularly for imaging modalities used between inpatient and outpatient groups, and in-hours versus out-of-hours.

Objectives: To study whether photon-counting computed tomography (PCCT) can improve CT number accuracy, precision and reduce patient size dependence compared to dual-energy CT (DECT) virtual monoenergetic imaging (VMI) and single-energy CT (SECT).

Methods: Clinical PCCT, DECT and SECT scanners were used to image a multi-energy quality assurance phantom and tissue-equivalent inserts with/without an outer nested annulus, representing two object sizes (18 and 33 cm). CT numbers were converted to linear attenuation coefficients (LAC) and regions-of-interest applied. Theoretical monoenergetic LAC were calculated from known elemental compositions as a ground truth. Percent difference in mean LAC between phantom sizes, between mean and theoretical LAC, and its coefficient of variation (COV) were calculated.

Results: Mean LAC percent differences between small and larger phantoms were highest in DECT (within -3% to 9%) and SECT (within 1% to 5%), particularly at higher calcium and iodine concentrations, while being relatively constant in PCCT over material concentrations and VMI energies (within ±2%). The COV in mean LAC was consistently lower (about 2 to 5 times) in PCCT relative to DECT and SECT for calcium in the large phantom. With consideration of the theoretical uncertainties of 2%, both PCCT and DECT showed comparable agreement to theoretical LAC.

Conclusions: PCCT VMI produces CT numbers with less dependence on patient size and increased precision in large object sizes than DECT VMI and SECT.

Advances in knowledge: Clinical PCCT provides less variable CT numbers than DECT and SECT with less sensitivity to the imaged object size.

Objectives: To assess the effect of plan complexity on dosimetric alterations induced by multi-leaf collimator (MLC) misplacements in volumetric modulated arc therapy (VMAT).

Methods: VMAT plans for 14 cervical and 10 lung cancer cases were reoptimized utilizing three distinct Aperture Shape Controller settings (none, very high, very low), resulting in three plan groups: ASC-none, ASC-vh, and ASC-vl. Four types of MLC position errors were simulated: total shifts (Type 1), open/closed (Type 2), right-side shifts (Type 3), and left-side shifts (Type 4). Plan complexity was assessed using the small aperture score (SAS). Dose deviations resulting from various MLC positioning errors and SAS values were calculated and compared among the three ASC groups.

Results: The variations in planning target volume (PTV) D95% for cervical cancer were approximately 0.6%, 3.7%, 1.9%, and 1.8% per millimeter for Types 1-4 errors, respectively. In the case of lung cancer, the changes were 2.3%, 9.3%, 5.3%, and 4.6% per millimeter. The ASC-vh and ASC-vl groups exhibited significantly reduced dose changes and SAS values in response to MLC errors, as compared to the ASC-none group (p < 0.05).

Conclusions: Highly complex plans exhibit greater dose sensitivity to MLC positional errors. The application of ASC proves effective in reducing plan complexity and mitigating the influence of MLC errors on dose deviation.

Advances in knowledge: By elucidating the relationship between dosimetric impacts from MLC errors and plan complexity, this study offers valuable guidance for the design of radiotherapy plans, helping to enhance the accuracy and effectiveness of VMAT treatments.