Journal of Medical Radiation Sciences最新文献

Maximise your continuing professional development (CPD) by reading the following selected article and answering the five questions. Please remember to self-claim your CPD and retain your supporting evidence. Answers will be available via the QR code and published in JMRS—Volume 72, Issue 4, December 2025.

Scan this QR code to find the answers.

Palliative radiation therapy (RT) is a vital treatment modality for managing symptoms from metastatic disease, but access barriers and workflow inefficiencies can delay or preclude care delivery. Simulation-free RT (SFRT) offers an effective, value-based solution by eliminating the traditional computed tomography (CT) simulation process and utilising existing diagnostic or staging CT scans for treatment planning. This process reduces patient burden and accelerates time to treatment, prioritising patient-centred care over traditional treatment pathways. Key technical considerations include managing dosimetric and geometric variations through appropriate patient selection and quality assurance processes. The successful implementation of SFRT requires a collaborative, multidisciplinary team approach, drawing on expertise from established centres to familiarise the team with the process. Access to diverse diagnostic imaging datasets and collaboration with various imaging providers is crucial. While careful patient selection is essential, our experience demonstrates that SFRT exemplifies value-based healthcare principles by optimising resource utilisation while prioritising patient-centred care, particularly valuable in rural settings where travel distances significantly impact treatment access. This paper aims to review the benefits and technical aspects, as well as provide key considerations for implementing SFRT in palliative RT settings.

The hypoglossal nerve (HN) provides motor innervation to tongue muscles responsible for tongue movement, speech, mastication, swallowing, respiratory functions and management of oral secretions. Injury, compression, entrapment or lesions of the HN at any point along its path can result in HN palsy and subsequent dysphagia, dysarthria and tongue muscular atrophy. A combined imaging approach is required to investigate the HN and causes of HN palsy. Magnetic resonance imaging (MRI) and computed tomography (CT) imaging are used to investigate the intracranial HN and where it emerges in the upper neck. The extracranial HN can be assessed by sonographic imaging along with the muscles directly and indirectly innervated by the HN. Ultrasound imaging can be challenging without an appropriate understanding of the detailed relative anatomy of the HN and the muscles it innervates, the associated sonographic technique and sonographic appearances, all of which are outlined in this paper.

Introduction: Chest X-rays (CXR) rank among the most conducted X-ray examinations. They often require repeat imaging due to inadequate quality, leading to increased radiation exposure and delays in patient care and diagnosis. This research assesses the efficacy of DenseNet121 and YOLOv8 neural networks in detecting suboptimal CXRs, which may minimise delays and enhance patient outcomes.

Method: The study included 3587 patients with a median age of 67 (0-102). It utilised an initial dataset comprising 10,000 CXRs randomly divided into a training subset (4000 optimal and 4000 suboptimal) and a validation subset (400 optimal and 400 suboptimal). The test subset (25 optimal and 25 suboptimal) was curated from the remaining images to provide adequate variation. Neural networks DenseNet121 and YOLOv8 were chosen due to their capabilities in image classification. DenseNet121 is a robust, well-tested model in the medical industry with high accuracy in object recognition. YOLOv8 is a cutting-edge commercial model targeted at all industries. Their performance was assessed via the area under the receiver operating curve (AUROC) and compared to radiologist classification, utilising the chi-squared test.

Results: DenseNet121 attained an AUROC of 0.97, while YOLOv8 recorded a score of 0.95, indicating a strong capability in differentiating between optimal and suboptimal CXRs. The alignment between radiologists and models exhibited variability, partly due to the lack of clinical indications. However, the performance was not statistically significant.

Conclusion: Both AI models effectively classified chest X-ray quality, demonstrating the potential for providing radiographers with feedback to improve image quality. Notably, this was the first study to include both PA and lateral CXRs as well as paediatric cases and the first to evaluate YOLOv8 for this application.