Large language models (LLMs) have shown potential in radiology, but their ability to aid radiologists in interpreting imaging studies remains unexplored. We investigated the effects of a state-of-the-...
In this retrospective study, six radiologists of different experience levels read 40 selected radiographic [n = 10], CT [n = 10], MRI [n = 10], and angiographic [n = 10] studies unassisted (session on...
When assessing if the correct diagnosis was among the top-3 differential diagnoses, diagnostic accuracy improved slightly from 181/240 (75.4%, unassisted) to 188/240 (78.3%, AI-assisted). Similar impr...
Integrating GPT-4 in the diagnostic process improved diagnostic accuracy slightly and diagnostic confidence significantly. Potentially harmful hallucinations and misinterpretations call for caution an...
Using GPT-4 as a virtual assistant when reading images made six radiologists of different experience levels feel more confident and provide more accurate diagnoses; yet, GPT-4 gave factually incorrect...
Medical doctors can encounter significant challenges in both the radiology image interpretation service and their ability to interpret images to promote effective patient management. This study aimed ...
A qualitative approach with a descriptive phenomenology design was employed. Thirteen medical officers and medical interns, with a maximum of three years of experience, were purposively selected from ...
Three main themes emerged from the data. The first theme was a poor image interpretation service which highlighted issues such as long turnaround times for image reporting and compromised patient mana...
Medical doctors in this low-resource setting experience significant delays in radiology image interpretation, leading to compromised patient management. Their training in image interpretation is inade...
There is a need to review and develop a comprehensive image interpretation system that effectively supports medical doctors in image interpretation, possibly involving the collaboration of radiographe...
To evaluate the degree of deformation in patients with ankle osteoarthritis (OA), it is essential to measure the three-dimensional (3D), in other words, stereoscopic alignment of the ankle, subtalar, ...
In this study, we compared the 2D human radiographic method with a stereoscopic image in patients with ankle arthritis. We enrolled 57 patients diagnosed with OA (28 left and 29 right) and obtained bo...
On the ICC between 2D radiographs and 3D analysis, the tibiotalar surface angle and lateral talo-1st metatarsal angle showed a good ICC grade (> 0.6), while other parameters did not have significant I...
We demonstrated that 2D and stereoscopic images are useful for the diagnosis of OA. Our study also confirmed that the radiographic features affected by ankle OA varied. However, according to the resul...
Dual-energy chest radiography exhibits better sensitivity than single-energy chest radiography, partly due to its ability to remove overlying anatomical structures....
To develop and validate a deep learning-based synthetic bone-suppressed (DLBS) nodule-detection algorithm for pulmonary nodule detection on chest radiographs....
This decision analytical modeling study used data from 3 centers between November 2015 and July 2019 from 1449 patients. The DLBS nodule-detection algorithm was trained using single-center data (insti...
DLBS nodule-detection algorithm....
The nodule-detection performance of DLBS model was compared with the convolution neural network nodule-detection algorithm (original model). Reader performance testing was conducted by 3 thoracic radi...
Training data consisted of 998 patients (539 men [54.0%]; mean [SD] age, 54.2 [9.82] years), and 2 external validation data sets consisted of 246 patients (133 men [54.1%]; mean [SD] age, 55.3 [8.7] y...
This decision analytical modeling study found that the DLBS model was more sensitive to detecting pulmonary nodules on chest radiographs compared with the original model. These findings suggest that t...
Lung cancer is typically classified into small-cell carcinoma and non-small-cell carcinoma. Non-small-cell carcinoma accounts for approximately 85% of all lung cancers. Low-dose chest computed tomogra...
Although the value of adding AI as a surrogate second reader in various scenarios has been investigated, it is unknown whether implementing an AI tool within double reading practice would capture addi...
To analyze dosimetric data of a single center by a radiation dose index monitoring software evaluating quantitatively the dose reduction obtained with the implementation of the adaptive statistical it...
Dosimetric quantities were acquired using Qaelum DOSE tool (QAELUM NV, Leuven-Heverlee, Belgium). Dose data pertaining to CT examinations were performed using a General Electric Healthcare CT tomograp...
Differences statistically significant were found for the DLP median values between the dosimetric data recorded on 2017-2018 versus 2019-2020. The differences were linked to the implementation of ASIR...
Automated methods of radiation dose data collection allowed for detailed radiation dose analysis according to protocol and equipment over time. The use of CT ASIR technique could determine considerabl...
What is known about checklists for interpreting chest radiographs? The question will guide the development of the inclusion criteria for the scoping review. Breaking down the scoping review question w...
X-ray reporting can be standardised using checklists. Checklists may reduce the time needed to produce a comprehensive X-ray report and improve the quality and consistency of detecting abnormalities o...
We will follow the methodological framework for scoping reviews originally described by Arksey and O'Malley. The scoping review will include articles that describe checklists for reducing diagnostic e...
The results will be collated, summarised and discussed including any limitations of the included articles. The articles will be summarised in a table, as well as narratively. The distribution of studi...
The results of the scoping review will be used to develop a checklist that will be used by medical doctors in collaboration with radiographers working in settings where there are no radiologists on-si...
Scoping review protocol registered with Open Science Framework on 27 July 2022. Registration https://doi.org/10.17605/OSF.IO/JS5PQ....
Background Supine chest radiography for bedridden patients in intensive care units (ICUs) is one of the most frequently ordered imaging studies worldwide. Purpose To evaluate the diagnostic performanc...
Ultrasonound is used to identify anatomical structures during regional anaesthesia and to guide needle insertion and injection of local anaesthetic. ScanNav Anatomy Peripheral Nerve Block (Intelligent...
Ultrasound-guided regional anaesthesia experts acquired 720 videos from 40 volunteers (across nine anatomical regions) without using the device. The artificial-intelligence colour overlay was subseque...
The artificial-intelligence models identified the structure of interest in 93.5% of cases (1519/1624), with a false-negative rate of 3.0% (48/1624) and a false-positive rate of 3.5% (57/1624). Highlig...
Artificial intelligence-based devices can potentially aid image acquisition and interpretation in ultrasound-guided regional anaesthesia. Further studies are necessary to demonstrate their effectivene...
NCT04906018....
