Expert Readers Research Articles

Modified PROMISE criteria for standardized interpretation of gastrin-releasing peptide receptor (GRPR)-targeted PET.

There are image interpretation criteria to standardize reporting prostate-specific membrane antigen (PSMA)-targeted positron emission tomography (PET). As up to 10% of prostate cancer (PC) do not express PSMA, other targets such as gastrin-releasing peptide receptor (GRPR) are evaluated. Research on GRPR-targeted imaging has been slowly increasing in usage at staging and biochemical recurrence (BCR) of PC. We therefore propose a modification of the Prostate Cancer Molecular Imaging Standardized Evaluation (PROMISE) criteria (mPROMISE) for GRPR-targeted PET. [68Ga]Ga-RM2 PET data from initially prospective studies performed at our institution were retrospectively reviewed: 44 patients were imaged for staging and 100 patients for BCR PC. Two nuclear medicine physicians independently evaluated PET according to the mPROMISE criteria. A third expert reader served as standard reference. Interreader reliability was computed for GRPR expression, prostate bed (T), lymph node (N), skeleton (Mb), organ (Mc) metastases, and final judgment of the scan. The interrater reliability for GRPR PET at staging was moderate for GRPR expression (0.59; 95% confidence interval [CI] 0.40, 0.78), substantial for T-stage (0.78; 95% CI 0.63, 0.94), and almost perfect for N-stage (0.97; 95% CI 0.92, 1.00) and final judgment (0.92; 95% CI 0.82, 1.00). The interreader agreement at BCR showed substantial agreement for GRPR expression (0.70; 95% CI 0.59, 0.81) and final judgment (0.65; 95% CI 0.53, 0.78), while almost perfect agreement was seen across the major categories (T, N, Mb, Mc). Acceptable performance of the mPROMISE criteria was found for all subsets when compared to the standard reference. Interpreting GRPR-targeted PET using the mPROMISE criteria showed its reliability with substantial or almost perfect interrater agreement across all major categories. The proposed modification of the PROMISE criteria will aid clinicians in decreasing the level of uncertainty, and clinical trials to achieve uniform evaluation, reporting, and comparability of GRPR-targeted PET. Clinicaltrials.gov Identifier: NCT03113617 and NCT02624518.

Convolutional Neural Networks to Assess Steno-Occlusive Disease Using Cerebrovascular Reactivity.

Cerebrovascular Reactivity (CVR) is a provocative test used with Blood oxygenation level-dependent (BOLD) Magnetic Resonance Imaging (MRI) studies, where a vasoactive stimulus is applied and the corresponding changes in the cerebral blood flow (CBF) are measured. The most common clinical application is the assessment of cerebral perfusion insufficiency in patients with steno-occlusive disease (SOD). Globally, millions of people suffer from cerebrovascular diseases, and SOD is the most common cause of ischemic stroke. Therefore, CVR analyses can play a vital role in early diagnosis and guiding clinical treatment. This study develops a convolutional neural network (CNN)-based clinical decision support system to facilitate the screening of SOD patients by discriminating between healthy and unhealthy CVR maps. The networks were trained on a confidential CVR dataset with two classes: 68 healthy control subjects, and 163 SOD patients. This original dataset was distributed in a ratio of 80%-10%-10% for training, validation, and testing, respectively, and image augmentations were applied to the training and validation sets. Additionally, some popular pre-trained networks were imported and customized for the objective classification task to conduct transfer learning experiments. Results indicate that a customized CNN with a double-stacked convolution layer architecture produces the best results, consistent with expert clinical readings.

Sense and nonsense of yT-staging on MRI after chemoradiotherapy in rectal cancer.

The aim of this work was to investigate the value of rectal cancer T-staging on MRI after chemoradiotherapy (ymrT-staging) in relation to the degree of fibrotic transformation of the tumour bed as assessed using the pathological tumour regression grade (pTRG) of Mandard as a standard of reference. Twenty two radiologists, including five rectal MRI experts and 17 'nonexperts' (general/abdominal radiologists), evaluated the ymrT stage on the restaging MRIs of 90 rectal cancer patients after chemoradiotherapy. The ymrT stage was compared with the final ypT stage at histopathology; the percentages of correct staging (ymrT = ypT), understaging (ymrT < ypT) and overstaging (ymrT > ypT) were calculated and compared between patients with predominant tumour at histopathology (pTRG4-5) and patients with predominant fibrosis (pTRG1-3). Interobserver agreement (IOA) was computed using Krippendorff's alpha. Average ymrT/ypT stage concordance was 48% for the experts and 43% for the nonexperts; ymrT/ypT stage concordance was significantly higher in the pTRG4-5 subgroup (58% vs. 41% for the pTRG1-3 group; p = 0.01), with the best results for the MRI experts. Overstaging was the main source of error, especially in the pTRG1-3 subgroup (average overstaging rate 38%-44% vs. 13%-55% in the pTRG4-5 subgroup). IOA was higher for the expert versus nonexpert readers (α = 0.67 vs. α = 0.39). ymrT-staging is moderately accurate; accuracy is higher in poorly responding patients with predominant tumour but low in good responders with predominant fibrosis, resulting in significant overstaging. Radiologists should shift their focus from ymrT-staging to detecting gross residual (and progressive) disease, and identifying potential candidates for organ preservation who would benefit from further clinical and endoscopic evaluation to guide final treatment planning.

Mapping the Spatial Extent of Hypoperfusion in Chronic Thromboembolic Pulmonary Hypertension Using Multienergy CT.

To assess if a novel automated method to spatially delineate and quantify the extent of hypoperfusion on multienergy CT angiograms can aid the evaluation of chronic thromboembolic pulmonary hypertension (CTEPH) disease severity. Multienergy CT angiograms obtained between January 2018 and December 2020 in 51 patients with CTEPH (mean age, 47 years ± 17 [SD]; 27 women) were retrospectively compared with those in 110 controls with no imaging findings suggestive of pulmonary vascular abnormalities (mean age, 51 years ± 16; 81 women). Parenchymal iodine values were automatically isolated using deep learning lobar lung segmentations. Low iodine concentration was used to delineate areas of hypoperfusion and calculate hypoperfused lung volume (HLV). Receiver operating characteristic curves, correlations with preoperative and postoperative changes in invasive hemodynamics, and comparison with visual assessment of lobar hypoperfusion by two expert readers were evaluated. Global HLV correctly separated patients with CTEPH from controls (area under the receiver operating characteristic curve = 0.84; 10% HLV cutoff: 90% sensitivity, 72% accuracy, and 64% specificity) and correlated moderately with hemodynamic severity at time of imaging (pulmonary vascular resistance [PVR], ρ = 0.67; P < .001) and change after surgical treatment (∆PVR, ρ = -0.61; P < .001). In patients surgically classified as having segmental disease, global HLV correlated with preoperative PVR (ρ = 0.81) and postoperative ∆PVR (ρ = -0.70). Lobar HLV correlated moderately with expert reader lobar assessment (ρHLV = 0.71 for reader 1; ρHLV = 0.67 for reader 2). Automated quantification of hypoperfused areas in patients with CTEPH can be performed from clinical multienergy CT examinations and may aid clinical evaluation, particularly in patients with segmental-level disease.Keywords: CT-Spectral Imaging (Multienergy), Pulmonary, Pulmonary Arteries, Embolism/Thrombosis, Chronic Thromboembolic Pulmonary Hypertension, Multienergy CT, Hypoperfusion© RSNA, 2023.

Machine learning based model to diagnose obstructive coronary artery disease using calcium scoring, PET imaging, and clinical data

IntroductionAccurate risk stratification in patients with suspected stable coronary artery disease is essential for choosing an appropriate treatment strategy. Our aim was to develop and validate a machine learning (ML) based model to diagnose obstructive CAD (oCAD). MethodWe retrospectively have included 1007 patients without a prior history of CAD who underwent CT-based calcium scoring (CACS) and a Rubidium-82 PET scan. The entire dataset was split 4:1 into a training and test dataset. An ML model was developed on the training set using fivefold stratified cross-validation. The test dataset was used to compare the performance of expert readers to the model. The primary endpoint was oCAD on invasive coronary angiography (ICA). ResultsROC curve analysis showed an AUC of 0.92 (95% CI 0.90-0.94) for the training dataset and 0.89 (95% CI 0.84-0.93) for the test dataset. The ML model showed no significant differences as compared to the expert readers (p ≥ 0.03) in accuracy (89% vs. 88%), sensitivity (68% vs. 69%), and specificity (92% vs. 90%). ConclusionThe ML model resulted in a similar diagnostic performance as compared to expert readers, and may be deployed as a risk stratification tool for obstructive CAD. This study showed that utilization of ML is promising in the diagnosis of obstructive CAD.

Early EEG-burst sharpness and 2-year disability in extremely preterm infants

BackgroundAutomated computational measures of EEG have the potential for large-scale application. We hypothesised that a predefined measure of early EEG-burst shape (increased burst sharpness) could predict neurodevelopmental impairment (NDI) and mental developmental index (MDI) at 2 years of age over-and-above that of brain ultrasound.MethodsWe carried out a secondary analysis of data from extremely preterm infants collected for an RCT (SafeBoosC-II). Two hours of single-channel cross-brain EEG was used to analyse burst sharpness with an automated algorithm. The co-primary outcomes were moderate-or-severe NDI and MDI. Complete data were available from 58 infants. A predefined statistical analysis was adjusted for GA, sex and no, mild–moderate, and severe brain injury as detected by cranial ultrasound.ResultsNine infants had moderate-or-severe NDI and the mean MDI was 87 ± 17.3 SD. The typical burst sharpness was low (negative values) and varied relatively little (mean –0.81 ± 0.11 SD), but the odds ratio for NDI was increased by 3.8 (p = 0.008) and the MDI was reduced by –3.2 points (p = 0.14) per 0.1 burst sharpness units increase (+1 SD) in the adjusted analysis.ConclusionThis study confirms the association between EEG-burst measures in preterm infants and neurodevelopment in childhood. Importantly, this was by a priori defined analysis.ImpactA fully automated, computational measure of EEG in the first week of life was predictive of neurodevelopmental impairment at 2 years of age.This confirms many previous studies using expert reading of EEG.Only single-channel EEG data were used, adding to the applicability.EEG was recorded by several different devices thus this measure appears to be robust to differences in electrodes, amplifiers and filters.The likelihood ratio of a positive EEG test, however, was only about 2, suggesting little immediate clinical value.

Assessing the role of an artificial intelligence assessment tool for thoracic aorta diameter on routine chest CT.

To assess the diagnostic accuracy and clinical impact of automated artificial intelligence (AI) measurement of thoracic aorta diameter on routine chest CT. A single-centre retrospective study involving three cohorts. 210 consecutive ECG-gated CT aorta scans (mean age 75 ± 13) underwent automated analysis (AI-Rad Companion Chest CT, Siemens) and were compared to a reference standard of specialist cardiothoracic radiologists for accuracy measuring aortic diameter. A repeated measures analysis tested reporting consistency in a second cohort (29 patients, mean age 61 ± 17) of immediate sequential pre-contrast and contrast CT aorta acquisitions. Potential clinical impact was assessed in a third cohort of 197 routine CT chests (mean age 66 ± 15) to document potential clinical impact. AI analysis produced a full report in 387/436 (89%) and a partial report in 421/436 (97%). Manual vs AI agreement was good to excellent (ICC 0.76-0.92). Repeated measures analysis of expert and AI reports for the ascending aorta were moderate to good (ICC 0.57-0.88). AI diagnostic performance crossed the threshold for maximally accepted limits of agreement (>5 mm) at the aortic root on ECG-gated CTs. AI newly identified aortic dilatation in 27% of patients on routine thoracic imaging with a specificity of 99% and sensitivity of 77%. AI has good agreement with expert readers at the mid-ascending aorta and has high specificity, but low sensitivity, at detecting dilated aortas on non-dedicated chest CTs. An AI tool may improve the detection of previously unknown thoracic aorta dilatation on chest CTs vs current routine reporting.

Integrating biomarkers and multi-parametric MRI to provide enhanced clinical diagnosis for prostate cancer

Prostate cancer (PCa) risk assessment can incorporate clinical features, gene expression, protein ‘biomarkers’ or imaging. In this review the benefits of layering multiparametric magnetic resonance imaging (mpMRI) with other risk assessment methods is considered. mpMRI is an increasingly utilized risk assessment tool in prostate cancer. The European Association of Urology, National Comprehensive Cancer Network (NCCN) and American Urological Association (AUA) guidelines call for mpMRI utilization in the prostate cancer management pathway. As such, the NCCN Guidelines and AUA guidelines emphasize differing levels of reliance on mpMRI preceding prostate biopsy. However, like all risk assessment tools, mpMRI has strengths and limitations. This include dependencies on reader expertise and interpretation, equipment and process standardization, tumor size, tumor multifocality, tissue architecture, ethnic and racial disparity, and cost. Thus, layering complementary risk assessment methods to mitigate the limitations of each approach, enables the most informed clinical management. The goal of ongoing biomarker/mpMRI studies is to provide insight into the clinically helpful integration of the two approaches. For new technologies to be adapted or layered together synergistically, five specific competencies must be considered acceptable: (1) efficacy, (2) potential side effect levels, (3) ease of use of technology, (4) cost vs. clinical benefit, and (5) durability.

Estimation of brain amyloid accumulation using deep learning in clinical [11C]PiB PET imaging

IntroductionEstimation of brain amyloid accumulation is valuable for evaluation of patients with cognitive impairment in both research and clinical routine. The development of high throughput and accurate strategies for the determination of amyloid status could be an important tool in patient selection for clinical trials and amyloid directed treatment. Here, we propose the use of deep learning to quantify amyloid accumulation using standardized uptake value ratio (SUVR) and classify amyloid status based on their PET images.MethodsA total of 1309 patients with cognitive impairment scanned with [11C]PIB PET/CT or PET/MRI were included. Two convolutional neural networks (CNNs) for reading-based amyloid status and SUVR prediction were trained using 75% of the PET/CT data. The remaining PET/CT (n = 300) and all PET/MRI (n = 100) data was used for evaluation.ResultsThe prevalence of amyloid positive patients was 61%. The amyloid status classification model reproduced the expert reader’s classification with 99% accuracy. There was a high correlation between reference and predicted SUVR (R2 = 0.96). Both reference and predicted SUVR had an accuracy of 97% compared to expert classification when applying a predetermined SUVR threshold of 1.35 for binary classification of amyloid status.ConclusionThe proposed CNN models reproduced both the expert classification and quantitative measure of amyloid accumulation in a large local dataset. This method has the potential to replace or simplify existing clinical routines and can facilitate fast and accurate classification well-suited for a high throughput pipeline.

Eastern High Arctic–Baffin Bay beluga whale (Delphinapterus leucas) estuary abundance and use from space

IntroductionThe Eastern High Arctic–Baffin Bay (EHA-BB) beluga whale (Delphinapterus leucas) population spends summer in estuaries around Somerset Island, Nunavut, Canada. A single abundance estimate from 1996 suggests an abundance &amp;gt;21,000 beluga whales; however, more information on abundance and distribution is needed to ensure effective management of this population, especially in estuaries where previous surveys provided minimal coverage. To assess the feasibility of using Very High Resolution (VHR) satellite imagery to obtain estuary abundance estimates for this beluga population, we evaluated a citizen science crowd counting initiative that was designed to monitor remote beluga whale populations and their estuary use.MethodsIn July and August 2020 the WorldView 2 and 3, and GeoEye 1 satellites were tasked to collect VHR imagery (30–41 cm) of estuaries previously known to be used by Eastern High Arctic–Baffin Bay beluga whales. The objectives were to obtain an estuary abundance estimate for this population from satellite imagery, and to evaluate the effectiveness of having imagery annotated using a crowd-source platform. Almost 3,800 km2of ocean imagery was analyzed using Maxar’s Geospatial Human Imagery Verification Effort (GeoHIVE) Crowdsourcing platform. Expert readers then manually compared counts to those performed by crowd-counters to determine variance in observer counts.Results and DiscussionThe estuary abundance estimate from 11 core estuaries was 12,128 (CV 36.76%, 95% confidence interval 6,036–24,368) beluga whales. This represents an estuary abundance estimate only, as the greater Peel Sound and Prince Regent Inlet areas were not photographed. The estuaries with the largest abundance of beluga whales were Creswell Bay, Maxwell Bay, and Prince Whales Island, with over 2,000 crowd-counted whales in each estuary. Although VHR imagery has potential to assist with surveying and monitoring marine mammals, for larger estuaries it was not always possible to photograph the entire area in a single day, and cloud cover was an issue for sections of most images. This work will assist with planning large-scale aerial surveys for monitoring beluga whale populations, identifying high-use areas and important beluga habitat, and highlights the utility of using VHR imagery to enhance our understanding of estuary abundance and distribution of Arctic whales.

Artificial intelligence-assisted interpretation of systolic function by echocardiogram

ObjectivePrecise and reliable echocardiographic assessment of left ventricular ejection fraction (LVEF) is needed for clinical decision-making. Recently, artificial intelligence (AI) models have been developed to estimate LVEF accurately. The aim...

Can Low-Iodine, Low-Radiation-Dose CT Aortogram Reliably Detect Endoleak after Endovascular Aneurysm Repair of the Aorta?

Double-low CT aortography (DLCTA) is increasingly used in follow-up studies of aortic aneurysm after endovascular aneurysm repair (EVAR). However, whether DLCTA can reliably detect the presence of endoleak is not clear. From February 2014 to October 2019, patients who received EVAR, underwent CT surveillance, and had at least one standard CTA protocol (120 kVp, 400 mg I/kg) and one DLCTA (70-80 kVp, 200 mg I/kg) were included. The integrated findings of the standard CTA and sequential change were considered as the reference standard for the presence of endoleak. In all, 36 patients received TEVAR and 24 patients received EVAR; 62 standard CTA and 167 DLCTA results were analyzed. There were 2 type I (3.3%) and 12 type II (20.0%) endoleaks in 14 patients (23.3%). The performance of DLCTA in the diagnosis of endoleak reached 100% accuracy compared to that of standard CTA in case of the correction of CT findings by an expert second reading. Compared to the standard CTA, DLCTA scan reduced the radiation dose by 71% and the iodine dose by 50%. DLCTA with 70-80 kVp and 200 mg I/kg can reliably detect the presence of endoleak after TEVAR/EVAR.

Comparison of the accuracy and reproducibility of AI based echocardiography and CMR analysis for cardiac function assessment in oncology patients at risk of cardiotoxicity

Abstract Funding Acknowledgements Type of funding sources: None. Background Serial screening for cancer therapy related cardiac dysfunction (CTRCD) demands precise measurement of left ventricular ejection fraction (LVEF) and global longitudinal strain (GLS) to detect cardiotoxicity early and avoid inappropriate treatment breaks. Precision can be maximised by better image quality and more accurate analysis(1). Cardiovascular magnetic resonance (CMR) is gold standard for image acquisition(2,3), but the benefit of fully-automated AI-based analysis is unclear(4). Purpose To compare the precision (test-retest reproducibility) of AI-based analysis with human measurement for quantification of LVEF and GLS using echocardiography (TTE) and CMR. Methods Consented adult oncology patients underwent paired same day repeat scans using TTE and CMR (GE Vivid9 machine and Siemens 1.5T Aera) following EACVI protocols. Manual image analysis was performed by 2 blinded experts (ECHO-PAC and CVI42 software). TTE images were analysed by two FDA-approved fully-automated (unsupervised) AI packages (EchoAI1 and EchoAI2), and CMR images by one FDA-approved AI package (CMRAI1) and one validated fully-automated inline deep-learning algorithm (CMRAI2). Precision was calculated between repeat studies for each modality and analysis technique by the mean absolute difference (MAD), co-efficient of variation (CoV), minimal detectable difference (MDD) and Bland-Altman limits of agreement; compared using Levene’s test. Results 48 patients (median age 52 years, 81% female, 60% breast cancer), underwent same day repeat TTE and CMR (4 scans). Manual median LVEF was lower with TTE than CMR 57% (95%CI 56.2–57.8) vs 61.4% (95%CI 61–67.7), p&amp;lt;0.01. Analysis failed in 8%, 12% and 4% TTE scans using EchoAI1, EchoAI2 and manual techniques. There were no significant differences in the MAD, CoV or MDD in measurements for LVEF (p&amp;gt;0.05) between analysis methods, Table 1. Human analysis was superior to EchoAI1 for measurements of GLS; MAD 1.37 vs 2.71 (p&amp;lt;0.01), MDD 2.5 vs 5.7 (p&amp;lt;0.01) and CoV 6.9% vs 10.6%, (p = 0.03). There were zero analysis failures for analysis of CMR data. The CoV (3.6% vs 4.5%, p = 0.03), MAD (1.8 vs 2.9, p = 0.03) and MDD (4.3 vs 4.8, p&amp;gt;0.5) for LVEF metrics was lower with CMRAI2 compared to human analysis. The CoV (6.2% vs 11.7%, P&amp;lt;0.01), MAD (1.23 vs 1.92, p = 0.03) and MDD (2.4 vs 3.6, p&amp;gt;0.05) for repeat GLS measurements was lower with human analysis compared with CMRAI1. There were no other significant differences, Table 1. Conclusions Measurement precision is higher with CMR than TTE, in keeping with prior studies. Fully automated AI algorithms can provide incremental improvements in LVEF measurement precision for analysing CMR images. AI techniques for analysis of echocardiographic images provide similar precision to expert human readers. With continued development, fully-automated analysis algorithms should improve confidence in CTRCD surveillance imaging for cancer patients, with results generalisable to other clinical indications.

Deep learning for assessing image quality in bi-parametric prostate MRI: A feasibility study

BackgroundAlthough systems such as Prostate Imaging Quality (PI-QUAL) have been proposed for quality assessment, visual evaluations by human readers remain somewhat inconsistent, particularly among less-experienced readers. ObjectivesTo assess the feasibility of deep learning (DL) for the automated assessment of image quality in bi-parametric MRI scans and compare its performance to that of less-experienced readers. MethodsWe used bi-parametric prostate MRI scans from the PI-CAI dataset in this study. A 3-point Likert scale, consisting of poor, moderate, and excellent, was utilized for assessing image quality. Three expert readers established the ground-truth labels for the development (500) and testing sets (100). We trained a 3D DL model on the development set using probabilistic prostate masks and an ordinal loss function. Four less-experienced readers scored the testing set for performance comparison. ResultsThe kappa scores between the DL model and the expert consensus for T2W images and ADC maps were 0.42 and 0.61, representing moderate and good levels of agreement. The kappa scores between the less-experienced readers and the expert consensus for T2W images and ADC maps ranged from 0.39 to 0.56 (fair to moderate) and from 0.39 to 0.62 (fair to good). ConclusionsDeep learning (DL) can offer performance comparable to that of less-experienced readers when assessing image quality in bi-parametric prostate MRI, making it a viable option for an automated quality assessment tool. We suggest that DL models trained on more representative datasets, annotated by a larger group of experts, could yield reliable image quality assessment and potentially substitute or assist visual evaluations by human readers.

Inter-reader agreement of the prostate imaging quality (PI-QUAL) score for basic readers in prostate MRI: A multi-center study

BackgroundThe Prostate Imaging Quality (PI-QUAL) score is the first step toward image quality assessment in multi-parametric prostate MRI (mpMRI). Previous studies have demonstrated moderate to excellent inter-rater agreement among expert readers; however, there is a need for studies to assess the inter-reader agreement of PI-QUAL scoring in basic prostate readers. ObjectivesTo assess the inter-reader agreement of the PI-QUAL score amongst basic prostate readers on multi-center prostate mpMRI. MethodsFive basic prostate readers from different centers assessed the PI-QUAL scores independently using T2-weighted images, diffusion-weighted imaging (DWI) including apparent diffusion coefficient (ADC) maps, and dynamic-contrast-enhanced (DCE) images on mpMRI data obtained from five different centers following Prostate Imaging-Reporting and Data System Version 2.1. The inter-reader agreements amongst radiologists for PI-QUAL were evaluated using weighted Cohen’s kappa. Further, the absolute agreements in assessing the diagnostic adequacy of each mpMRI sequence were calculated. ResultsA total of 355 men with a median age of 71 years (IQR, 60–78) were enrolled in the study. The pair-wise kappa scores ranged from 0.656 to 0.786 for the PI-QUAL scores, indicating good inter-reader agreements between the readers. The pair-wise absolute agreements ranged from 0.75 to 0.88 for T2W imaging, from 0.74 to 0.83 for the ADC maps, and from 0.77 to 0.86 for DCE images. ConclusionsBasic prostate radiologists from different institutions provided good inter-reader agreements on multi-center data for the PI-QUAL scores.

Development of a bespoke phantom to optimize molecular PET imaging of pituitary tumors

BackgroundImage optimization is a key step in clinical nuclear medicine, and phantoms play an essential role in this process. However, most phantoms do not accurately reflect the complexity of human anatomy, and this presents a particular challenge when imaging endocrine glands to detect small (often subcentimeter) tumors. To address this, we developed a novel phantom for optimization of positron emission tomography (PET) imaging of the human pituitary gland. Using radioactive 3D printing, phantoms were created which mimicked the distribution of 11C-methionine in normal pituitary tissue and in a small tumor embedded in the gland (i.e., with no inactive boundary, thereby reproducing the in vivo situation). In addition, an anatomical phantom, replicating key surrounding structures [based on computed tomography (CT) images from an actual patient], was created using material extrusion 3D printing with specialized filaments that approximated the attenuation properties of bone and soft tissue.ResultsThe phantom enabled us to replicate pituitary glands harboring tumors of varying sizes (2, 4 and 6 mm diameters) and differing radioactive concentrations (2 ×, 5 × and 8 × the normal gland). The anatomical phantom successfully approximated the attenuation properties of surrounding bone and soft tissue. Two iterative reconstruction algorithms [ordered subset expectation maximization (OSEM); Bayesian penalized likelihood (BPL)] with a range of reconstruction parameters (e.g., 3, 5, 7 and 9 OSEM iterations with 24 subsets; BPL regularization parameter (β) from 50 to 1000) were tested. Images were analyzed quantitatively and qualitatively by eight expert readers. Quantitatively, signal was the highest using BPL with β = 50; noise was the lowest using BPL with β = 1000; contrast was the highest using BPL with β = 100. The qualitative review found that accuracy and confidence were the highest when using BPL with β = 400.ConclusionsThe development of a bespoke phantom has allowed the identification of optimal parameters for molecular pituitary imaging: BPL reconstruction with TOF, PSF correction and a β value of 400; in addition, for small (< 4 mm) tumors with low contrast (2:1 or 5:1), sensitivity may be improved using a β value of 100. Together, these findings should increase tumor detection and confidence in reporting scans.

Artificial Intelligence Can Generate Fraudulent but Authentic-Looking Scientific Medical Articles: Pandora's Box Has Been Opened.

Artificial intelligence (AI) has advanced substantially in recent years, transforming many industries and improving the way people live and work. In scientific research, AI can enhance the quality and efficiency of data analysis and publication. However, AI has also opened up the possibility of generating high-quality fraudulent papers that are difficult to detect, raising important questions about the integrity of scientific research and the trustworthiness of published papers. The aim of this study was to investigate the capabilities of current AI language models in generating high-quality fraudulent medical articles. We hypothesized that modern AI models can create highly convincing fraudulent papers that can easily deceive readers and even experienced researchers. This proof-of-concept study used ChatGPT (Chat Generative Pre-trained Transformer) powered by the GPT-3 (Generative Pre-trained Transformer 3) language model to generate a fraudulent scientific article related to neurosurgery. GPT-3 is a large language model developed by OpenAI that uses deep learning algorithms to generate human-like text in response to prompts given by users. The model was trained on a massive corpus of text from the internet and is capable of generating high-quality text in a variety of languages and on various topics. The authors posed questions and prompts to the model and refined them iteratively as the model generated the responses. The goal was to create a completely fabricated article including the abstract, introduction, material and methods, discussion, references, charts, etc. Once the article was generated, it was reviewed for accuracy and coherence by experts in the fields of neurosurgery, psychiatry, and statistics and compared to existing similar articles. The study found that the AI language model can create a highly convincing fraudulent article that resembled a genuine scientific paper in terms of word usage, sentence structure, and overall composition. The AI-generated article included standard sections such as introduction, material and methods, results, and discussion, as well a data sheet. It consisted of 1992 words and 17 citations, and the whole process of article creation took approximately 1 hour without any special training of the human user. However, there were some concerns and specific mistakes identified in the generated article, specifically in the references. The study demonstrates the potential of current AI language models to generate completely fabricated scientific articles. Although the papers look sophisticated and seemingly flawless, expert readers may identify semantic inaccuracies and errors upon closer inspection. We highlight the need for increased vigilance and better detection methods to combat the potential misuse of AI in scientific research. At the same time, it is important to recognize the potential benefits of using AI language models in genuine scientific writing and research, such as manuscript preparation and language editing.

Development of a new radiographic score for the follow-up of calcific tendinopathy of the rotator cuff.

The aim of this study was to evaluate the psychometric properties of a new x-ray scoring system for calcific tendinopathy of the rotator cuff (CTRC). This is a post-hoc analysis of the CALCECHO trial. All patients received an ultrasound-guided puncture and lavage of their calcification. Clinical data and x-rays from baseline and follow-up visits at 7 days (D7), 3 months (M3) and 12 months (M12) were used. The scoring system was based on the reduction in size and density of the calcification compared to the initial x-ray (0 = no change; 1 = decrease of less than 50%; 2 = decrease of between 50 and 90%; 3 = decrease of more than 90%; 4 = complete disappearance). Inter-observer and intra-observer reliability were established between 3 independent investigators (2 experts and one junior) using weighted Kappa calculation. Construct validity was assessed as well as predictive validity and sensitivity to change. Between the two experts, inter-reader reliability was at 0.677, 0.744 and 0.656 at D7, M3 and M12 respectively. Intra-reader reliability was between 0.577 and 0.836 for the two expert readers and between 0.519 and 0.697 for the junior reader. Our score was correlated with shoulder pain and function at M3 and M12 and the score at M3 was predictive of the clinical outcome at M12. Finally, sensitivity to change was 0.8. Our new score presented good psychometric properties and was correlated with clinical data. It could be useful in the follow-up of patients treated for CTRC.

Comparison of CT and Dixon MR Abdominal Adipose Tissue Quantification Using a Unified Computer-Assisted Software Framework.

Reliable and objective measures of abdominal fat distribution across imaging modalities are essential for various clinical and research scenarios, such as assessing cardiometabolic disease risk due to obesity. We aimed to compare quantitative measures of subcutaneous (SAT) and visceral (VAT) adipose tissues in the abdomen between computed tomography (CT) and Dixon-based magnetic resonance (MR) images using a unified computer-assisted software framework. This study included 21 subjects who underwent abdominal CT and Dixon MR imaging on the same day. For each subject, two matched axial CT and fat-only MR images at the L2-L3 and the L4-L5 intervertebral levels were selected for fat quantification. For each image, an outer and an inner abdominal wall regions as well as SAT and VAT pixel masks were automatically generated by our software. The computer-generated results were then inspected and corrected by an expert reader. There were excellent agreements for both abdominal wall segmentation and adipose tissue quantification between matched CT and MR images. Pearson coefficients were 0.97 for both outer and inner region segmentation, 0.99 for SAT, and 0.97 for VAT quantification. Bland-Altman analyses indicated minimum biases in all comparisons. We showed that abdominal adipose tissue can be reliably quantified from both CT and Dixon MR images using a unified computer-assisted software framework. This flexible framework has a simple-to-use workflow to measure SAT and VAT from both modalities to support various clinical research applications.

Reading numbers is harder than reading words: An eye-tracking study

We recorded the eye movements of adults reading aloud short (four digit) and long (eight to 11 digit) Arabic numerals compared to matched-in-length words and pseudowords. We presented each item in isolation, at the center of the screen. Participants read each item aloud at their pace, and then pressed the spacebar to display the next item. Reading accuracy was 99 %. Results showed that adults make 2.5 times more fixations when reading short numerals compared to short words, and up to 7 times more fixations when reading long numerals with respect to long words. Similarly, adults make 3 times more saccades when reading short numerals compared to short words, and up to 9 times more saccades when reading long numerals with respect to long words. Fixation duration and saccade amplitude stay almost the same when reading short numerals with respect to short words. However, fixation duration increases by ∼50 ms when reading long numerals (∼300 ms) with respect to long words (∼250 ms), and saccade amplitude decreases up to 0.83 characters when reading long numerals with respect to long words. The pattern of findings for long numerals—more and shorter saccades as well as more and longer fixations—shows the extent to which reading long Arabic numerals is a cognitively costly task. Within the phonographic writing system, this pattern of eye movements stands for the use of the sublexical print-to-sound correspondence rules. The data highlight that reading large numerals is an unautomatized activity and that Arabic numerals must be converted into their oral form by a step-by-step process even by expert readers.