FreshRSS

🔒
❌ Acerca de FreshRSS
Hay nuevos artículos disponibles. Pincha para refrescar la página.
AnteayerTus fuentes RSS

Evaluating the performance of artificial intelligence software for lung nodule detection on chest radiographs in a retrospective real-world UK population

Por: Maiter · A. · Hocking · K. · Matthews · S. · Taylor · J. · Sharkey · M. · Metherall · P. · Alabed · S. · Dwivedi · K. · Shahin · Y. · Anderson · E. · Holt · S. · Rowbotham · C. · Kamil · M. A. · Hoggard · N. · Balasubramanian · S. P. · Swift · A. · Johns · C. S.
Objectives

Early identification of lung cancer on chest radiographs improves patient outcomes. Artificial intelligence (AI) tools may increase diagnostic accuracy and streamline this pathway. This study evaluated the performance of commercially available AI-based software trained to identify cancerous lung nodules on chest radiographs.

Design

This retrospective study included primary care chest radiographs acquired in a UK centre. The software evaluated each radiograph independently and outputs were compared with two reference standards: (1) the radiologist report and (2) the diagnosis of cancer by multidisciplinary team decision. Failure analysis was performed by interrogating the software marker locations on radiographs.

Participants

5722 consecutive chest radiographs were included from 5592 patients (median age 59 years, 53.8% women, 1.6% prevalence of cancer).

Results

Compared with radiologist reports for nodule detection, the software demonstrated sensitivity 54.5% (95% CI 44.2% to 64.4%), specificity 83.2% (82.2% to 84.1%), positive predictive value (PPV) 5.5% (4.6% to 6.6%) and negative predictive value (NPV) 99.0% (98.8% to 99.2%). Compared with cancer diagnosis, the software demonstrated sensitivity 60.9% (50.1% to 70.9%), specificity 83.3% (82.3% to 84.2%), PPV 5.6% (4.8% to 6.6%) and NPV 99.2% (99.0% to 99.4%). Normal or variant anatomy was misidentified as an abnormality in 69.9% of the 943 false positive cases.

Conclusions

The software demonstrated considerable underperformance in this real-world patient cohort. Failure analysis suggested a lack of generalisability in the training and testing datasets as a potential factor. The low PPV carries the risk of over-investigation and limits the translation of the software to clinical practice. Our findings highlight the importance of training and testing software in representative datasets, with broader implications for the implementation of AI tools in imaging.

Characterising the interventions designed to affect the reporting of musculoskeletal imaging: a scoping review protocol using the COM-B model

Por: Kirby · E. · MacMillan · A. · Liew · B. X. W. · Brinkley · A. · Bateman · A.
Introduction

Attributing musculoskeletal (MSK) pain to normal and commonly occurring imaging findings, such as tendon, cartilage and spinal disc degeneration, has been shown to increase people’s fear of movement, reduce their optimism about recovery and increase healthcare costs. Interventions seeking to reduce the negative effects of MSK imaging reporting have had little effect. To understand the ineffectiveness of these interventions, this study seeks to scope their behavioural targets, intended mechanisms of action and theoretical underpinnings. This information alongside known barriers to helpful reporting can enable researchers to refine or create new more targeted interventions.

Methods and analysis

The scoping review will be conducted in accordance with the JBI methodology for scoping reviews and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews. Search terms will be devised by the research team. Searches of MEDLINE, EMBASE, CINAHL, AMED and PsycINFO from inception to current day will be performed. The review will include studies, which have developed or evaluated interventions targeting the reporting of MSK imaging. Studies targeting the diagnosis of serious causes of MSK pain will be excluded. Two independent authors will extract study participant data using predefined extraction templates and intervention details using the Template for Intervention Description and Replication checklist. Interventions will be coded and mapped to the technique, mechanism of action and behavioural target according to the Capability, Opportunity, Motivation-Behaviour (COM-B) model categories. Any explicit models or theories used to inform the selection of interventions will be extracted and coded. The study characteristics, behaviour change techniques identified, behavioural targets according to the COM-B and context specific theories within the studies will be presented in narrative and table form.

Ethics and dissemination

The information from this review will be used to inform an intervention design process seeking to improve the communication of imaging results. The results will also be disseminated through a peer-reviewed publication, conference presentations and stakeholder events.

AI assisted reader evaluation in acute CT head interpretation (AI-REACT): protocol for a multireader multicase study

Por: Fu · H. · Novak · A. · Robert · D. · Kumar · S. · Tanamala · S. · Oke · J. · Bhatia · K. · Shah · R. · Romsauerova · A. · Das · T. · Espinosa · A. · Grzeda · M. T. · Narbone · M. · Dharmadhikari · R. · Harrison · M. · Vimalesvaran · K. · Gooch · J. · Woznitza · N. · Salik · N. · Campbell · A.
Introduction

A non-contrast CT head scan (NCCTH) is the most common cross-sectional imaging investigation requested in the emergency department. Advances in computer vision have led to development of several artificial intelligence (AI) tools to detect abnormalities on NCCTH. These tools are intended to provide clinical decision support for clinicians, rather than stand-alone diagnostic devices. However, validation studies mostly compare AI performance against radiologists, and there is relative paucity of evidence on the impact of AI assistance on other healthcare staff who review NCCTH in their daily clinical practice.

Methods and analysis

A retrospective data set of 150 NCCTH will be compiled, to include 60 control cases and 90 cases with intracranial haemorrhage, hypodensities suggestive of infarct, midline shift, mass effect or skull fracture. The intracranial haemorrhage cases will be subclassified into extradural, subdural, subarachnoid, intraparenchymal and intraventricular. 30 readers will be recruited across four National Health Service (NHS) trusts including 10 general radiologists, 15 emergency medicine clinicians and 5 CT radiographers of varying experience. Readers will interpret each scan first without, then with, the assistance of the qER EU 2.0 AI tool, with an intervening 2-week washout period. Using a panel of neuroradiologists as ground truth, the stand-alone performance of qER will be assessed, and its impact on the readers’ performance will be analysed as change in accuracy (area under the curve), median review time per scan and self-reported diagnostic confidence. Subgroup analyses will be performed by reader professional group, reader seniority, pathological finding, and neuroradiologist-rated difficulty.

Ethics and dissemination

The study has been approved by the UK Healthcare Research Authority (IRAS 310995, approved 13 December 2022). The use of anonymised retrospective NCCTH has been authorised by Oxford University Hospitals. The results will be presented at relevant conferences and published in a peer-reviewed journal.

Trial registration number

NCT06018545.

Sex-specific Impact of the first COVID-19 Lockdown on Age Structure and Case Acuity at Admission in a Patient Population in southwestern Germany: a retrospective comparative Study in Neuroradiology

Por: Reder · S. R. · Herrlich · N. · Grauhan · N. F. · Othman · A. E. · Müller-Eschner · M. · Brockmann · C. · Brockmann · M. A.
Objectives

A hard lockdown was presumed to lead to delayed diagnosis and treatment of serious diseases, resulting in higher acuity at admission. This should be elaborated based on the estimated acuity of the cases, changes in findings during hospitalisation, age structure and biological sex.

Design

Retrospective monocentric cross-sectional study.

Setting

German Neuroradiology Department at a .

Participants

In 2019, n=1158 patients were admitted in contrast to n=884 during the first hard lockdown in 2020 (11th–13th week).

Main outcome measures

Three radiologists evaluated the initial case acuity, classified them into three groups (not acute, subacute and acute), and evaluated if there was a relevant clinical deterioration. The data analysis was conducted using non-parametric methods and multivariate regression analysis.

Results

A 24% decrease in the number of examinations from 2019 to 2020 (p=0.025) was revealed. In women, the case acuity increased by 21% during the lockdown period (p=0.002). A 30% decrease in acute cases in men was observable (in women 5% decrease). Not acute cases decreased in both women and men (47%; 24%), while the subacute cases remained stable in men (0%) and decreased in women (28%). Regression analysis revealed the higher the age, the higher the acuity (p

Conclusion

The lockdown led to a decrease in neuroradiological consultations, with delays in seeking medical care. In women, the number of most severe cases remained stable, whereas the mean case acuity and age increased. This could be due to greater pandemic-related anxiety among women, however, with severe symptoms they were seeking for medical help. In contrast in men, the absolute number of most severe cases decreased, whereas the mean acuity and age remained nearly unaffected. This could be attributable to a reduced willingness to seek for medical consultation.

Developing blood-brain barrier arterial spin labelling as a non-invasive early biomarker of Alzheimers disease (DEBBIE-AD): a prospective observational multicohort study protocol

Por: Padrela · B. · Mahroo · A. · Tee · M. · Sneve · M. H. · Moyaert · P. · Geier · O. · Kuijer · J. P. A. · Beun · S. · Nordhoy · W. · Zhu · Y. D. · Buck · M. A. · Hoinkiss · D. C. · Konstandin · S. · Huber · J. · Wiersinga · J. · Rikken · R. · de Leeuw · D. · Grydeland · H. · Tippett · L. · Caw
Introduction

Loss of blood-brain barrier (BBB) integrity is hypothesised to be one of the earliest microvascular signs of Alzheimer’s disease (AD). Existing BBB integrity imaging methods involve contrast agents or ionising radiation, and pose limitations in terms of cost and logistics. Arterial spin labelling (ASL) perfusion MRI has been recently adapted to map the BBB permeability non-invasively. The DEveloping BBB-ASL as a non-Invasive Early biomarker (DEBBIE) consortium aims to develop this modified ASL-MRI technique for patient-specific and robust BBB permeability assessments. This article outlines the study design of the DEBBIE cohorts focused on investigating the potential of BBB-ASL as an early biomarker for AD (DEBBIE-AD).

Methods and analysis

DEBBIE-AD consists of a multicohort study enrolling participants with subjective cognitive decline, mild cognitive impairment and AD, as well as age-matched healthy controls, from 13 cohorts. The precision and accuracy of BBB-ASL will be evaluated in healthy participants. The clinical value of BBB-ASL will be evaluated by comparing results with both established and novel AD biomarkers. The DEBBIE-AD study aims to provide evidence of the ability of BBB-ASL to measure BBB permeability and demonstrate its utility in AD and AD-related pathologies.

Ethics and dissemination

Ethics approval was obtained for 10 cohorts, and is pending for 3 cohorts. The results of the main trial and each of the secondary endpoints will be submitted for publication in a peer-reviewed journal.

❌