A Deep Learning-Based Radiomic Classifier for Usual Interstitial Pneumonia.

BACKGROUND: Because chest CT scan has largely supplanted surgical lung biopsy for diagnosing most cases of interstitial lung disease (ILD), tools to standardize CT scan interpretation are urgently needed. RESEARCH QUESTION: Does a deep learning (DL)-based classifier for usual interstitial pneumonia (UIP) derived using CT scan features accurately discriminate radiologist-determined visual UIP? STUDY DESIGN AND METHODS: A retrospective cohort study was performed. Chest CT scans acquired in individuals with and without ILD were drawn from a variety of public and private data sources. Using radiologist-determined visual UIP as ground truth, a convolutional neural network was used to learn discrete CT scan features of UIP, with outputs used to predict the likelihood of UIP using a linear support vector machine. Test performance characteristics were assessed in an independent performance cohort and multicenter ILD clinical cohort. Transplant-free survival was compared between UIP classification approaches using the Kaplan-Meier estimator and Cox proportional hazards regression. RESULTS: A total of 2,907 chest CT scans were included in the training (n = 1,934), validation (n = 408), and performance (n = 565) data sets. The prevalence of radiologist-determined visual UIP was 12.4% and 37.1% in the performance and ILD clinical cohorts, respectively. The DL-based UIP classifier predicted visual UIP in the performance cohort with sensitivity and specificity of 93% and 86%, respectively, and in the multicenter ILD clinical cohort with 81% and 77%, respectively. DL-based and visual UIP classification similarly discriminated survival, and outcomes were consistent among cases with positive DL-based UIP classification irrespective of visual classification. INTERPRETATION: A DL-based classifier for UIP demonstrated good test performance across a wide range of UIP prevalence and similarly discriminated survival when compared with radiologist-determined UIP. This automated tool could efficiently screen for UIP in patients undergoing chest CT scan and identify a high-risk phenotype among those with known ILD.

Standardising the organisation of clinical equipment on surgical wards at North Bristol NHS Trust: a quality improvement initiative.

Poorly organised clinical equipment can waste significant amounts of time otherwise available for direct patient care. As a group of foundation year one doctors, we identified the organisation of clinical equipment across surgical wards at North Bristol NHS Trust to be poor with stocks often low and items frequently difficult to locate. Time-motion studies (n=80) were confirmatory demonstrating that the mean time to collect equipment necessary for venepuncture, cannulation, arterial blood gases, or blood cultures ranged from 121 to 174 seconds between different areas. By applying a plan-do-study-act (PDSA) methodology, surveying peers as well as working with nursing staff and senior managers, we were able to purchase and implement clinical equipment trolleys on 10 surgical wards across the trust to reduce the time-taken to locate clinical equipment to between 38 to 45 seconds (p=0.01). We feel the key factors for the success of our initiative were strong multidisciplinary engagement and a simple uniform idea. Clinical equipment trolleys organised in a standardised manner have now been rolled out hospital-wide in the new Southmead Hospital Brunel building.

Engaging all doctors in continuous quality improvement: a structured, supported programme for first-year doctors across a training deanery in England.

The structure of postgraduate medical training rightly puts enormous emphasis on gathering clinical experience and constantly updating knowledge of relevant medical research to use in practice. At most, this can be contrasted with the slight emphasis on clinical leadership and acquiring the skills to effect change and improve the quality of care. Doctors play central roles in orchestrating the clinical management of patients across multiple settings within the healthcare system. They also routinely encounter the many problems within the systems that they work, affecting their own practices as well as those of other healthcare professionals. They thus represent a tremendous resource for identifying solutions to these problems and playing leadership roles in implementing them. However, physician training programs focus almost entirely on the knowledge and skills to manage clinical problems, with almost no training in skills related to healthcare management or effective quality improvement. In this article, we describe one attempt to improve this situation. In four hospitals in the Severn Deanery in the Southwest of England, first-year doctors carry out a structured and supported quality improvement project of their choice throughout their first year of training. To date, 30 such projects have been or are being run. This has significant benefits for both the trusts they are working for as well as for their own professional development. We describe the successes, difficulties and future of this programme.

The role of thoracic imaging in the intensive care unit.

Radiology is critical in managing patients on intensive care units. The portable chest X-ray is widely used, but ultrasound and computed tomography are of diagnostic value in selected cases. This article discusses the role of thoracic imaging in the intensive care unit.

Radiological investigation of acute stroke 1: non-enhanced computed tomography.

Radiology plays a central part in the acute stroke management pathway, with its role now widened beyond establishing the diagnosis. This article reviews the role of the non-enhanced computed tomography brain scan, particularly focusing on the hyper-acute presentation of stroke from a radiological perspective.

Radiological investigation of acute stroke 2: the multimodal approach.

Stroke is an important condition with high mortality and morbidity. Radiology plays a central part in the acute stroke management pathway, with its role now widened beyond establishing the diagnosis. This article reviews radiological techniques beyond the non-enhanced computed tomography brain scan.

The role of multi-detector computed tomography in imaging hernias.

Clinical presentation of a hernia is often non-specific or atypical and in these circumstances diagnosis and management decisions can be aided by imaging. This review contains diagrammatic illustrations, explanations and computed tomography examples of the different types of external, internal and diaphragmatic hernias.

Surgical management of single ostium anomalous coronary artery from pulmonary artery.

A 6-week-old infant presenting in heart failure was suspected on echocardiography of having anomalous origin of the whole coronary circulation from the pulmonary artery arising from a single ostium. This was confirmed at operation, and the child underwent successful correction. The management of cardiopulmonary bypass and myocardial protection in this abnormality is discussed.

Eight years survival after partial left ventriculectomy.

Remodelling the left ventricle by partial left ventriculectomy (PLV) has been proposed as an alternative surgical treatment for end-stage heart failure. We present the case of a patient with dilated idiopathic cardiomyopathy who underwent PLV and remains in NYHA functional Class I eight years after the surgery.