Explore the vast range of titles available.

ObjectiveChatGPT (Generative Pre-trained Transformer) is an artificial intelligence language tool developed by OpenAI that utilises machine learning algorithms to generate text that closely mimics human language. It has recently taken the internet by storm. There have been several concerns regarding the accuracy of documents it generates. This study compares the accuracy and quality of several ChatGPT-generated academic articles with those written by human authors.Material and methodsWe performed a study to assess the accuracy of ChatGPT-generated radiology articles by comparing them with the published or written, and under review articles. These were independently analysed by two fellowship-trained musculoskeletal radiologists and graded from 1 to 5 (1 being bad and inaccurate to 5 being excellent and accurate).ResultsIn total, 4 of the 5 articles written by ChatGPT were significantly inaccurate with fictitious references. One of the papers was well written, with a good introduction and discussion; however, all references were fictitious.ConclusionChatGPT is able to generate coherent research articles, which on initial review may closely resemble authentic articles published by academic researchers. However, all of the articles we assessed were factually inaccurate and had fictitious references. It is worth noting, however, that the articles generated may appear authentic to an untrained reader.

In 2017, I wrote an editorial for the Journal of Contemporary Administration (RAC), which contributed to advancing my thoughts on technological articles. After the publication, I received many more invitations from graduate programs, associations, scientific events, and the Brazilian agency Capes to discuss technological production, particularly technological articles. This experience highlighted two things for me: (a) after five years, it is time to update the thoughts I had in 2017; (b) being unfamiliar with this type of study is the main barrier for people to produce technological articles. Thus, I am writing the following lines to invite you to a new reflection on what a technological article is. However, it is important to recognize that, strictly speaking, a technological article is a scientific article and is not worse or better than a traditional one. Depending on the audience, it may be more or less relevant, but not worse or better.

ObjectiveTo compare performances in diagnosing intertrochanteric hip fractures from proximal femoral radiographs between a convolutional neural network and orthopedic surgeons.Materials and methodsIn total, 1773 patients were enrolled in this study. Hip plain radiographs from these patients were cropped to display only proximal fractured and non-fractured femurs. Images showing pseudarthrosis after femoral neck fracture and those showing artificial objects were excluded. This yielded a total of 3346 hip images (1773 fractured and 1573 non-fractured hip images) that were used to compare performances between the convolutional neural network and five orthopedic surgeons.ResultsThe convolutional neural network and orthopedic surgeons had accuracies of 95.5% (95% CI = 93.1–97.6) and 92.2% (95% CI = 89.2–94.9), sensitivities of 93.9% (95% CI = 90.1–97.1) and 88.3% (95% CI = 83.3–92.8), and specificities of 97.4% (95% CI = 94.5–99.4) and 96.8% (95% CI = 95.1–98.4), respectively.ConclusionsThe performance of the convolutional neural network exceeded that of orthopedic surgeons in detecting intertrochanteric hip fractures from proximal femoral radiographs under limited conditions. The convolutional neural network has a significant potential to be a useful tool for screening for fractures on plain radiographs, especially in the emergency room, where orthopedic surgeons are not readily available.

Abstract ObjectivesTo compare the value of zero echo time (ZTE) and gradient echo “black bone” (BB) MRI sequences for bone assessment of the sacroiliac joint (SI) using computed tomography (CT) as the reference standard.Materials and methodsBetween May 2019 and January 2021, 79 patients prospectively underwent clinically indicated 3-T MRI including ZTE and BB imaging. Additionally, all patients underwent a CT scan covering the SI joints within 12 months of the MRI examination. Two blinded readers performed bone assessment by grading each side of each SI joint qualitatively in terms of seven features (osteophytes, subchondral sclerosis, erosions, ankylosis, joint irregularity, joint widening, and gas in the SI joint) using a 4-point Likert scale (0 = no changes–3 = marked changes). Scores were compared between all three imaging modalities.ResultsInterreader agreement was largely good (k values: 0.5–0.83). Except for the feature “gas in SI joint” where ZTE exhibited significantly lower scores than CT (p < 0.001), ZTE and BB showed similar performance relative to CT for all other features (p > 0.52) with inter-modality agreement being substantial to almost perfect (Krippendorff’s alpha coefficients: 0.724–0.983). When combining the data from all features except for gas in the SI joint and when binarizing grading scores, combined sensitivity/specificity was 76.7%/98.6% for ZTE and 80.8%/99.1% for BB, respectively, compared to CT.ConclusionsThe performance of ZTE and BB sequences was comparable to CT for bone assessment of the SI joint. These sequences may potentially serve as an alternative to CT yet without involving exposure to ionizing radiation.

Background Despite clear assessment criteria, studies of molar incisor hypomineralisation (MIH) and hypomineralised second primary molars (HSPM) are marked by inconsistency in outcome measurements. This has detracted from meaningful comparisons between studies and limited interpretation. Aim To provide a comprehensive manual as a companion to assist researchers in planning epidemiological studies of MIH and HSPM, with particular reference to outcome measurement. Methods This manual begins with a succinct review of the clinical problems and evidence for management of the conditions. The subsequent sections guide researchers through diagnosis of MIH and HSPM and implementation of both the long and short forms of a recently proposed grading system. MIH and HSPM can often be confused with fluorosis, enamel hypoplasia, amelogenesis imperfecta, and white spot lesions but can be distinguished by a number of unique clinical features. Based on the grading system, a standardised protocol is proposed for clinical examinations. Intra and inter-examiner reliability is of key importance when outcome measurement is subjective and should be reported in all epidemiological studies of MIH. The manual concludes with an exercise forum aimed to train examiners in the use of the grading system, with answers provided. Conclusion The use of a standardised protocol, diagnostic and grading criteria will greatly enhance the quality of epidemiological studies of MIH.

ObjectiveThe study aims to evaluate the diagnostic performance of deep learning–based reconstruction method (DLRecon) in 3D MR neurography for assessment of the brachial and lumbosacral plexus.Materials and methodsThirty-five exams (18 brachial and 17 lumbosacral plexus) of 34 patients undergoing routine clinical MR neurography at 1.5 T were retrospectively included (mean age: 49 ± 12 years, 15 female). Coronal 3D T2-weighted short tau inversion recovery fast spin echo with variable flip angle sequences covering plexial nerves on both sides were obtained as part of the standard protocol. In addition to standard-of-care (SOC) reconstruction, k-space was reconstructed with a 3D DLRecon algorithm.Two blinded readers evaluated images for image quality and diagnostic confidence in assessing nerves, muscles, and pathology using a 4-point scale. Additionally, signal-to-noise ratio (SNR) and contrast-to-noise ratios (CNR) between nerve, muscle, and fat were measured.For comparison of visual scoring result non-parametric paired sample Wilcoxon signed-rank testing and for quantitative analysis paired sample Student’s t-testing was performed.ResultsDLRecon scored significantly higher than SOC in all categories of image quality (p < 0.05) and diagnostic confidence (p < 0.05), including conspicuity of nerve branches and pathology. With regard to artifacts there was no significant difference between the reconstruction methods.Quantitatively, DLRecon achieved significantly higher CNR and SNR than SOC (p < 0.05).ConclusionDLRecon enhanced overall image quality, leading to improved conspicuity of nerve branches and pathology, and allowing for increased diagnostic confidence in evaluation of the brachial and lumbosacral plexus.

ObjectiveTo develop and evaluate deep learning (DL) risk assessment models for predicting pain progression in subjects with or at risk of knee osteoarthritis (OA).Materials and methodsThe incidence and progression cohorts of the Osteoarthritis Initiative, a multi-center longitudinal study involving 9348 knees in 4674 subjects with or at risk of knee OA that began in 2004 and is ongoing, were used to conduct this retrospective analysis. A subset of knees without and with pain progression (defined as a 9-point or greater increase in pain score between baseline and two or more follow-up time points over the first 48 months) was randomly stratified into training (4200 knees with a mean age of 61.0 years and 60% female) and hold-out testing (500 knees with a mean age of 60.8 years and 60% female) datasets. A DL model was developed to predict pain progression using baseline knee radiographs. An artificial neural network was used to develop a traditional risk assessment model to predict pain progression using demographic, clinical, and radiographic risk factors. A combined model was developed to combine demographic, clinical, and radiographic risk factors with DL analysis of baseline knee radiographs. Area under the curve (AUC) analysis was performed using the hold-out testing dataset to evaluate model performance.ResultsThe traditional model had an AUC of 0.692 (66.9% sensitivity and 64.1% specificity). The DL model had an AUC of 0.770 (76.7% sensitivity and 70.5% specificity), which was significantly higher (p < 0.001) than the traditional model. The combined model had an AUC of 0.807 (72.3% sensitivity and 80.9% specificity), which was significantly higher (p < 0.05) than the traditional and DL models.ConclusionsDL models using baseline knee radiographs had higher diagnostic performance for predicting pain progression than traditional models using demographic, clinical, and radiographic risk factors.

ObjectiveTo compare rib fracture detection and classification by radiologists using CT images with and without a deep learning model.Materials and methodsA total of 8529 chest CT images were collected from multiple hospitals for training the deep learning model. The test dataset included 300 chest CT images acquired using a single CT scanner. The rib fractures were marked in the bone window on each CT slice by experienced radiologists, and the ground truth included 861 rib fractures. We proposed a heterogeneous neural network for rib fracture detection and classification consisting of a cascaded feature pyramid network and a classification network. The deep learning-based model was evaluated based on the external testing data. The precision rate, recall rate, F1-score, and diagnostic time of two junior radiologists with and without the deep learning model were computed, and the Chi-square, one-way analysis of variance, and least significant difference tests were used to analyze the results.ResultsThe use of the deep learning model increased detection recall and classification accuracy (0.922 and 0.863) compared with the radiologists alone (0.812 vs. 0.850). The radiologists achieved a higher precision rate, recall rate, and F1-score for fracture detection when using the deep learning model, at 0.943, 0.978, and 0.960, respectively. When using the deep learning model, the radiologist’s reading time was decreased from 158.3 ± 35.7 s to 42.3 ± 6.8 s.ConclusionRadiologists achieved the highest performance in diagnosing and classifying rib fractures on CT images when assisted by the deep learning model.

ObjectiveWe aimed to perform an external validation of an existing commercial AI software program (BoneView™) for the detection of acute appendicular fractures in pediatric patients.Materials and methodsIn our retrospective study, anonymized radiographic exams of extremities, with or without fractures, from pediatric patients (aged 2–21) were included. Three hundred exams (150 with fractures and 150 without fractures) were included, comprising 60 exams per body part (hand/wrist, elbow/upper arm, shoulder/clavicle, foot/ankle, leg/knee). The Ground Truth was defined by experienced radiologists. A deep learning algorithm interpreted the radiographs for fracture detection, and its diagnostic performance was compared against the Ground Truth, and receiver operating characteristic analysis was done. Statistical analyses included sensitivity per patient (the proportion of patients for whom all fractures were identified) and sensitivity per fracture (the proportion of fractures identified by the AI among all fractures), specificity per patient, and false-positive rate per patient.ResultsThere were 167 boys and 133 girls with a mean age of 10.8 years. For all fractures, sensitivity per patient (average [95% confidence interval]) was 91.3% [85.6, 95.3], specificity per patient was 90.0% [84.0,94.3], sensitivity per fracture was 92.5% [87.0, 96.2], and false-positive rate per patient in patients who had no fracture was 0.11. The patient-wise area under the curve was 0.93 for all fractures. AI diagnostic performance was consistently high across all anatomical locations and different types of fractures except for avulsion fractures (sensitivity per fracture 72.7% [39.0, 94.0]).ConclusionThe BoneView™ deep learning algorithm provides high overall diagnostic performance for appendicular fracture detection in pediatric patients.

Abstract ObjectiveTo compare the image quality of ultra-high-resolution wrist CTs acquired on photon-counting detector CT versus conventional energy-integrating-detector CT systems.Materials and methodsParticipants were scanned on a photon-counting-detector CT system after clinical energy-integrating detector CTs. Energy-integrating-detector CT scan parameters: comb filter-based ultra-high-resolution mode, 120 kV, 250 mAs, Ur70 or Ur73 kernel, 0.4- or 0.6-mm section thickness. Photon-counting-detector CT scan parameters: non-comb-based ultra-high-resolution mode, 120 kV, 120 mAs, Br84 kernel, 0.4-mm section thickness. Two musculoskeletal radiologists blinded to CT system, scored specific osseous structures using a 5-point Likert scale (1 to 5). The Wilcoxon rank-sum test was used for statistical analysis of reader scores. Paired t-test was used to compare volume CT dose index, bone CT number, and image noise between CT systems. P-value < 0.05 was considered statistically significant.ResultsTwelve wrists (mean participant age 55.3 ± 17.8, 6 females, 6 males) were included. The mean volume CT dose index was lower for photon-counting detector CT (9.6 ± 0.1 mGy versus 19.0 ± 6.7 mGy, p < .001). Photon-counting-detector CT images had higher Likert scores for visualization of osseous structures (median score = 4, p < 0.001). The mean bone CT number was higher in photon-counting-detector CT images (1946 ± 77 HU versus 1727 ± 49 HU, p < 0.001). Conversely, there was no difference in the mean image noise of the two CT systems (63 ± 6 HU versus 61 ± 6 HU, p = 0.13).ConclusionUltra-high-resolution imaging with photon-counting-detector CT depicted wrist structures more clearly than conventional energy-integrating-detector CT despite a 49% radiation dose reduction.