BJR open最新文献

Objectives: Accurate beam modelling is essential for dose calculation in stereotactic radiation therapy (SRT), such as CyberKnife treatment. However, the present deep learning methods only involve patient anatomical images and delineated masks for training. These studies generally focus on traditional intensity-modulated radiation therapy (RT) plans. Nevertheless, this paper aims to develop a deep CNN-based method for CyberKnife plan dose prediction about brain cancer patients. It utilized modelled beam information, target delineation, and patient anatomical information.

Methods: This study proposes a method that adds beam information to predict the dose distribution of CyberKnife in brain cases. A retrospective dataset of 88 brain and abdominal cancer patients treated with the Ray-tracing algorithm was performed. The datasets include patients' anatomical information (planning CT), binary masks for organs at risk (OARs) and targets, and clinical plans (containing beam information). The datasets were randomly split into 68, 6, and 14 brain cases for training, validation, and testing, respectively.

Results: Our proposed method performs well in SRT dose prediction. First, for the gamma passing rates in brain cancer cases, with the 2 mm/2% criteria, we got 96.7% ± 2.9% for the body, 98.3% ± 3.0% for the planning target volume, and 100.0% ± 0.0% for the OARs with small volumes referring to the clinical plan dose. Secondly, the model predictions matched the clinical plan's dose-volume histograms reasonably well for those cases. The differences in key metrics at the target area were generally below 1.0 Gy (approximately a 3% difference relative to the prescription dose).

Conclusions: The preliminary results for selected 14 brain cancer cases suggest that accurate 3-dimensional dose prediction for brain cancer in CyberKnife can be accomplished based on accurate beam modelling for homogeneous tumour tissue. More patients and other cancer sites are needed in a further study to validate the proposed method fully.

Advances in knowledge: With accurate beam modelling, the deep learning model can quickly generate the dose distribution for CyberKnife cases. This method accelerates the RT planning process, significantly improves its operational efficiency, and optimizes it.

Large language models (LLMs) are transforming the field of natural language processing (NLP). These models offer opportunities for radiologists to make a meaningful impact in their field. NLP is a part of artificial intelligence (AI) that uses computer algorithms to study and understand text data. Recent advances in NLP include the Attention mechanism and the Transformer architecture. Transformer-based LLMs, such as GPT-4 and Gemini, are trained on massive amounts of data and generate human-like text. They are ideal for analysing large text data in academic research and clinical practice in radiology. Despite their promise, LLMs have limitations, including their dependency on the diversity and quality of their training data and the potential for false outputs. Albeit these limitations, the use of LLMs in radiology holds promise and is gaining momentum. By embracing the potential of LLMs, radiologists can gain valuable insights and improve the efficiency of their work. This can ultimately lead to improved patient care.

Metabolic imaging in clinical practice has long relied on PET with fluorodeoxyglucose (FDG), a radioactive tracer. However, this conventional method presents inherent limitations such as exposure to ionizing radiation and potential diagnostic uncertainties, particularly in organs with heightened glucose uptake like the brain. This review underscores the transformative potential of traditional deuterium MR spectroscopy (MRS) when integrated with gradient techniques, culminating in an advanced metabolic imaging modality known as deuterium MRI (DMRI). While recent advancements in hyperpolarized MRS hold promise for metabolic analysis, their widespread clinical usage is hindered by cost constraints and the availability of hyperpolarizer devices or facilities. DMRI, also denoted as deuterium metabolic imaging (DMI), represents a pioneering, single-shot, and noninvasive paradigm that fuses conventional MRS with nonradioactive deuterium-labelled substrates. Extensively tested in animal models and patient cohorts, particularly in cases of brain tumours, DMI's standout feature lies in its seamless integration into standard clinical MRI scanners, necessitating only minor adjustments such as radiofrequency coil tuning to the deuterium frequency. DMRI emerges as a versatile tool for quantifying crucial metabolites in clinical oncology, including glucose, lactate, glutamate, glutamine, and characterizing IDH mutations. Its potential applications in this domain are broad, spanning diagnostic profiling, treatment response monitoring, and the identification of novel therapeutic targets across diverse cancer subtypes.

Objectives: To analyse changes in the use of paediatric (≤16 years) CT over the past decade and to evaluate the appropriateness of CT examinations at a tertiary teaching hospital.

Methods: Data from 290 paediatric CTs were prospectively collected in 2022 and compared with data from 2017 (358 cases) and 2012 (538 cases). The justification of CTs was evaluated with regard to medical imaging referral guidelines and appropriateness rates were calculated.

Results: Paediatric CTs decreased 39.4% over the 10 years, contrasting with a 27.6% increase in overall CTs. Paediatric CTs as the share of overall CTs dropped from 2.5% in 2012 to 1.1% in 2022 (P < .0001), with a concurrent rise in paediatric MRIs (P < .0001). Notable reductions in CT use occurred for head trauma (P = .0003), chronic headache (P < .0001), epilepsy (P = .037), hydrocephalus (P = .0078), chest tumour (P = .0005), and whole-body tumour (P = .0041). The overall appropriateness of CTs improved from 73.1% in 2017 to 79.0% in 2022 (P = .0049). In 15.4% of the cases, no radiological examination was deemed necessary, and in 8.7% of the cases, another modality was more appropriate. Appropriateness rates were the highest for the head and neck angiography (100%) and the chest (96%) and the lowest for the neck (66%) and the head (67%).

Conclusions: Justification of CT scans can be improved by regular educational interventions, increasing MRI accessibility, and evaluating the appropriateness of the requested CT before the examination. Interventions for a more effective implementation of referral guidelines are needed.

Advances in knowledge: The focus for improvement should be CTs for head and cervical spine trauma, accounting for the majority of inappropriate requests in the paediatric population.

Recent advancements in CT technology have introduced a revolutionary innovation to practice known as the Photon-Counting detector (PCD) CT imaging. The pivotal hardware enhancement of the PCD-CT scanner lies in its detectors, which consist of smaller pixels than standard detectors and allow direct conversion of individual X-rays to electrical signals. As a result, CT images are reconstructed at higher spatial resolution (as low as 0.2 mm) and reduced overall noise, at no expense of an increased radiation dose. These features are crucial for paediatric imaging, especially for infants and young children, where anatomical structures are notably smaller than in adults and in whom keeping dose as low as possible is especially relevant. Since January 2022, our hospital has had the opportunity to work with PCD-CT technology for paediatric imaging. This pictorial review will showcase clinical examples of PCD-CT imaging in children. The aim of this pictorial review is to outline the potential paediatric applications of PCD-CT across different anatomical regions, as well as to discuss the benefits in utilizing PCD-CT in comparison to conventional standard energy integrating detector CT.

Objectives: We validated an auto-contouring algorithm for heart substructures in lung cancer patients, aiming to establish its accuracy and reliability for radiotherapy (RT) planning. We focus on contouring an amalgamated set of subregions in the base of the heart considered to be a new organ at risk, the cardiac avoidance area (CAA), to enable maximum dose limit implementation in lung RT planning.

Methods: The study validates a deep-learning model specifically adapted for auto-contouring the CAA (which includes the right atrium, aortic valve root, and proximal segments of the left and right coronary arteries). Geometric, dosimetric, quantitative, and qualitative validation measures are reported. Comparison with manual contours, including assessment of interobserver variability, and robustness testing over 198 cases are also conducted.

Results: Geometric validation shows that auto-contouring performance lies within the expected range of manual observer variability despite being slightly poorer than the average of manual observers (mean surface distance for CAA of 1.6 vs 1.2 mm, dice similarity coefficient of 0.86 vs 0.88). Dosimetric validation demonstrates consistency between plans optimized using auto-contours and manual contours. Robustness testing confirms acceptable contours in all cases, with 80% rated as "Good" and the remaining 20% as "Useful."

Conclusions: The auto-contouring algorithm for heart substructures in lung cancer patients demonstrates acceptable and comparable performance to human observers.

Advances in knowledge: Accurate and reliable auto-contouring results for the CAA facilitate the implementation of a maximum dose limit to this region in lung RT planning, which has now been introduced in the routine setting at our institution.

Recent advances in percutaneous image-guided techniques have empowered interventional radiologists with diverse treatment options for the management of musculoskeletal lesions. Of note, there is growing utility for cementoplasty procedures, with indications ranging from stabilization of bone metastases to treatment of painful vertebral compression fractures. Likewise, cryoablation has emerged as a viable adjunct in the treatment of both primary and secondary bone and soft tissue neoplasms. These treatment options have been progressively incorporated into the multidisciplinary approach to holistic care of patients, alongside conventional radiotherapy, systemic therapy, surgery, and analgesia. This review article serves to outline the indications, technical considerations, latest developments, and evidence for the burgeoning role of cementoplasty and cryoablation in the musculoskeletal system, with an emphasis on pain palliation and tumour control.

[This corrects the article DOI: 10.1093/bjro/tzad005.].

Objectives: This study compared plans of high definition (HD), 2.5 mm width multi-leaf collimator (MLC), to standard, 5 mm width, isocentric linear accelerator (linacs), CyberKnife (CK), and Gamma Knife (GK) for stereotactic radiosurgery (SRS) techniques on multiple brain metastases.

Methods: Eleven patients undergoing SRS for multiple brain metastases were chosen. Targets and organs at risk (OARs) were delineated and optimized SRS plans were generated and compared.

Results: The linacs delivered similar conformity index (CI) values, but the gradient index (GI) for HD MLCs was significantly lower (P-value <.001). Half the OARs received significantly lower dose using HD MLCs. CK delivered a significantly lower CI than HD MLC linac (P-value <.001), but a significantly higher GI (P-value <.001). CI was significantly improved with the HD MLC linac compared to GK (P-value = 4.591 × 10^-3), however, GK delivered a significantly lower GI (P-value <.001). OAR dose sparing was similar for the HD MLC TL, CK, and GK.

Conclusions: Comparing linacs for SRS, the preferred choice is HD MLCs. Similar results were achieved with the HD MLC linac, CK, or GK, with each delivering significant improvements in different aspects of plan quality.

Advances in knowledge: This article is the first to compare HD and standard width MLC linac plans using a combination of single isocentre volumetric modulated arc therapy and multi-isocentric dynamic conformal arc plans as required, which is a more clinically relevant assessment. Furthermore, it compares these plans with CK and GK, assessing the relative merits of each technique.

Objectives: This diagnostic study assessed the accuracy of radiologists retrospectively, using the deep learning and natural language processing chest algorithms implemented in Clinical Review version 3.2 for: pneumothorax, rib fractures in digital chest X-ray radiographs (CXR); aortic aneurysm, pulmonary nodules, emphysema, and pulmonary embolism in CT images.

Methods: The study design was double-blind (artificial intelligence [AI] algorithms and humans), retrospective, non-interventional, and at a single NHS Trust. Adult patients (≥18 years old) scheduled for CXR and CT were invited to enroll as participants through an opt-out process. Reports and images were de-identified, processed retrospectively, and AI-flagged discrepant findings were assigned to two lead radiologists, each blinded to patient identifiers and original radiologist. The radiologist's findings for each clinical condition were tallied as a verified discrepancy (true positive) or not (false positive).

Results: The missed findings were: 0.02% rib fractures, 0.51% aortic aneurysm, 0.32% pulmonary nodules, 0.92% emphysema, and 0.28% pulmonary embolism. The positive predictive values (PPVs) were: pneumothorax (0%), rib fractures (5.6%), aortic dilatation (43.2%), pulmonary emphysema (46.0%), pulmonary embolus (11.5%), and pulmonary nodules (9.2%). The PPV for pneumothorax was nil owing to lack of available studies that were analysed for outpatient activity.

Conclusions: The number of missed findings was far less than generally predicted. The chest algorithms deployed retrospectively were a useful quality tool and AI augmented the radiologists' workflow.

Advances in knowledge: The diagnostic accuracy of our radiologists generated missed findings of 0.02% for rib fractures CXR, 0.51% for aortic dilatation, 0.32% for pulmonary nodule, 0.92% for pulmonary emphysema, and 0.28% for pulmonary embolism for CT studies, all retrospectively evaluated with AI used as a quality tool to flag potential missed findings. It is important to account for prevalence of these chest conditions in clinical context and use appropriate clinical thresholds for decision-making, not relying solely on AI.