ABSTRACT
Background: Primary CNS lymphoma (PCNSL) and glioblastoma (GBM) both represent frequent intracranial malignancies with differing clinical management. However, distinguishing PCNSL from GBM with conventional MRI can be challenging when atypical imaging features are present. We employed advanced dMRI for noninvasive characterization of the microstructure of PCNSL and differentiation from GBM as the most frequent primary brain malignancy. Methods: Multiple dMRI metrics including Diffusion Tensor Imaging, Neurite Orientation Dispersion and Density Imaging, and Diffusion Microstructure Imaging were extracted from the contrast-enhancing tumor component in 10 PCNSL and 10 age-matched GBM on 3T MRI. Imaging findings were correlated with cell density and axonal markers obtained from histopathology. Results: We found significantly increased intra-axonal volume fractions (V-intra and intracellular volume fraction) and microFA in PCNSL compared to GBM (all Pâ <â .001). In contrast, mean diffusivity (MD), axial diffusivity (aD), and microADC (all Pâ <â .001), and also free water fractions (V-CSF and V-ISO) were significantly lower in PCNSL (all Pâ <â .01). Receiver-operating characteristic analysis revealed high predictive values regarding the presence of a PCNSL for MD, aD, microADC, V-intra, ICVF, microFA, V-CSF, and V-ISO (area under the curve [AUC] in all >0.840, highest for MD and ICVF with an AUC of 0.960). Comparative histopathology between PCNSL and GBM revealed a significantly increased cell density in PCNSL and the presence of axonal remnants in a higher proportion of samples. Conclusions: Advanced diffusion imaging enables the characterization of the microstructure of PCNSL and reliably distinguishes PCNSL from GBM. Both imaging and histopathology revealed a relatively increased cell density and a preserved axonal microstructure in PCNSL.
ABSTRACT
PURPOSE: CT perfusion (CTP) is a valuable tool in suspected acute ischemic stroke. A substantial variability of the delay between contrast injection and bolus arrival in the brain is conceivable. We investigated the distribution of the peak positions of the concentration time curves measured in an artery (arterial input function, AIF) and - in cases with ischemia - also measured in the penumbra. METHODS: We report on 2624 perfusion scans (52 % female, mean age 72.2 ± 14.4 years) with stroke present in 1636 cases. From the attenuation time curves of the AIF and the penumbra, we calculated the respective bolus peak positions and investigated the distribution of the peak positions. Further, we analyzed the bolus peak positions for associations with age. RESULTS: The bolus peaked significantly later in older patients, both in the AIF and in the penumbra (all p < 0.001). In the whole cohort, we found a significant association of age with the bolus peak position of the AIF (ρ = 0.334; p < 0.001). In patients with stroke, age was also associated to the peak position of the AIF (ρ = 0.305; p < 0.001), and the penumbra (ρ = 0.246, p < 0.001). However, a substantial range of peak positions of the AIF and penumbra was noted across all age ranges. CONCLUSIONS: This study revealed a strong age-dependency of the contrast bolus arrival in both healthy and ischemic tissue. This variability makes non-uniform sampling schemes, which have been suggested to reduce radiation dose, problematic, as they might not always optimally capture the bolus in all cases.
Subject(s)
Contrast Media , Humans , Female , Male , Aged , Tomography, X-Ray Computed/methods , Aged, 80 and over , Ischemic Stroke/diagnostic imaging , Age Factors , Middle Aged , Reproducibility of Results , Cohort Studies , Stroke/diagnostic imagingABSTRACT
Cerebral intraparenchymal hemorrhage due to electrode implantation (CIPHEI) is a rare but serious complication of deep brain stimulation (DBS) surgery. This study retrospectively investigated a large single-center cohort of DBS implantations to calculate the frequency of CIPHEI and identify patient- and procedure-related risk factors for CIPHEI and their potential interactions. We analyzed all DBS implantations between January 2013 and December 2021 in a generalized linear model for binomial responses using bias reduction to account for sparse sampling of CIPHEIs. As potential risk factors, we considered age, gender, history of arterial hypertension, level of invasivity, types of micro/macroelectrodes, and implanted DBS electrodes. If available, postoperative coagulation and platelet function were exploratorily assessed in CIPHEI patients. We identified 17 CIPHEI cases across 839 electrode implantations in 435 included procedures in 418 patients (3.9%). Exploration and cross-validation analyses revealed that the three-way interaction of older age (above 60 years), high invasivity (i.e., use of combined micro/macroelectrodes), and implantation of directional DBS electrodes accounted for 82.4% of the CIPHEI cases. Acquired platelet dysfunction was present only in one CIPHEI case. The findings at our center suggested implantation of directional DBS electrodes as a new potential risk factor, while known risks of older age and high invasivity were confirmed. However, CIPHEI risk is not driven by the three factors alone but by their combined presence. The contributions of the three factors to CIPHEI are hence not independent, suggesting that potentially modifiable procedural risks should be carefully evaluated when planning DBS surgery in patients at risk.
ABSTRACT
OBJECTIVE: To investigate whether choroid plexus volumes in subacute coronavirus disease 2019 (COVID-19) patients with neurological symptoms could indicate inflammatory activation or barrier dysfunction and assess their association with clinical data. METHODS: Choroid plexus volumes were measured in 28 subacute COVID-19 patients via cerebral magnetic resonance imaging (MRI), compared with those in infection-triggered non-COVID-19 encephalopathy patients (n = 25), asymptomatic individuals after COVID-19 (n = 21), and healthy controls (n = 21). Associations with inflammatory serum markers (peak counts of leukocytes, C-reactive protein [CRP], interleukin 6), an MRI-based marker of barrier dysfunction (CSF volume fraction [V-CSF]), and clinical parameters like olfactory performance and cognitive scores (Montreal Cognitive Assessment) were investigated. RESULTS: COVID-19 patients showed significantly larger choroid plexus volumes than control groups (p < 0.001, η2 = 0.172). These volumes correlated significantly with peak leukocyte levels (p = 0.001, Pearson's r = 0.621) and V-CSF (p = 0.009, Spearman's rho = 0.534), but neither with CRP nor interleukin 6. No significant correlations were found with clinical parameters. INTERPRETATION: In patients with subacute COVID-19, choroid plexus volume is a marker of central nervous system inflammation and barrier dysfunction in the presence of neurologic symptoms. The absence of plexus enlargement in infection-triggered non-COVID-19 encephalopathy suggests a specific severe acute respiratory syndrome coronavirus 2 effect. This study also documents an increase in choroid plexus volume for the first time as a parainfectious event. ANN NEUROL 2024.
ABSTRACT
After contracting COVID-19, a substantial number of individuals develop a Post-COVID-Condition, marked by neurologic symptoms such as cognitive deficits, olfactory dysfunction, and fatigue. Despite this, biomarkers and pathophysiological understandings of this condition remain limited. Employing magnetic resonance imaging, we conduct a comparative analysis of cerebral microstructure among patients with Post-COVID-Condition, healthy controls, and individuals that contracted COVID-19 without long-term symptoms. We reveal widespread alterations in cerebral microstructure, attributed to a shift in volume from neuronal compartments to free fluid, associated with the severity of the initial infection. Correlating these alterations with cognition, olfaction, and fatigue unveils distinct affected networks, which are in close anatomical-functional relationship with the respective symptoms.
Subject(s)
COVID-19 , Cognitive Dysfunction , Fatigue , Magnetic Resonance Imaging , Olfaction Disorders , SARS-CoV-2 , Humans , COVID-19/complications , COVID-19/diagnostic imaging , COVID-19/physiopathology , COVID-19/pathology , Cognitive Dysfunction/diagnostic imaging , Cognitive Dysfunction/physiopathology , Cognitive Dysfunction/virology , Male , Fatigue/physiopathology , Female , Middle Aged , Olfaction Disorders/diagnostic imaging , Olfaction Disorders/virology , Olfaction Disorders/physiopathology , Adult , Brain/diagnostic imaging , Brain/pathology , Brain/physiopathology , Post-Acute COVID-19 Syndrome , AgedABSTRACT
BACKGROUND: We investigated the potential of an imaging-aware GPT-4-based chatbot in providing diagnoses based on imaging descriptions of abdominal pathologies. METHODS: Utilizing zero-shot learning via the LlamaIndex framework, GPT-4 was enhanced using the 96 documents from the Radiographics Top 10 Reading List on gastrointestinal imaging, creating a gastrointestinal imaging-aware chatbot (GIA-CB). To assess its diagnostic capability, 50 cases on a variety of abdominal pathologies were created, comprising radiological findings in fluoroscopy, MRI, and CT. We compared the GIA-CB to the generic GPT-4 chatbot (g-CB) in providing the primary and 2 additional differential diagnoses, using interpretations from senior-level radiologists as ground truth. The trustworthiness of the GIA-CB was evaluated by investigating the source documents as provided by the knowledge-retrieval mechanism. Mann-Whitney U test was employed. RESULTS: The GIA-CB demonstrated a high capability to identify the most appropriate differential diagnosis in 39/50 cases (78%), significantly surpassing the g-CB in 27/50 cases (54%) (p = 0.006). Notably, the GIA-CB offered the primary differential in the top 3 differential diagnoses in 45/50 cases (90%) versus g-CB with 37/50 cases (74%) (p = 0.022) and always with appropriate explanations. The median response time was 29.8 s for GIA-CB and 15.7 s for g-CB, and the mean cost per case was $0.15 and $0.02, respectively. CONCLUSIONS: The GIA-CB not only provided an accurate diagnosis for gastrointestinal pathologies, but also direct access to source documents, providing insight into the decision-making process, a step towards trustworthy and explainable AI. Integrating context-specific data into AI models can support evidence-based clinical decision-making. RELEVANCE STATEMENT: A context-aware GPT-4 chatbot demonstrates high accuracy in providing differential diagnoses based on imaging descriptions, surpassing the generic GPT-4. It provided formulated rationale and source excerpts supporting the diagnoses, thus enhancing trustworthy decision-support. KEY POINTS: ⢠Knowledge retrieval enhances differential diagnoses in a gastrointestinal imaging-aware chatbot (GIA-CB). ⢠GIA-CB outperformed the generic counterpart, providing formulated rationale and source excerpts. ⢠GIA-CB has the potential to pave the way for AI-assisted decision support systems.
Subject(s)
Artificial Intelligence , Gastrointestinal Diseases , Proof of Concept Study , Humans , Diagnosis, Differential , Gastrointestinal Diseases/diagnostic imagingABSTRACT
tRNA modifications affect ribosomal elongation speed and co-translational folding dynamics. The Elongator complex is responsible for introducing 5-carboxymethyl at wobble uridine bases (cm5U34) in eukaryotic tRNAs. However, the structure and function of human Elongator remain poorly understood. In this study, we present a series of cryo-EM structures of human ELP123 in complex with tRNA and cofactors at four different stages of the reaction. The structures at resolutions of up to 2.9 Å together with complementary functional analyses reveal the molecular mechanism of the modification reaction. Our results show that tRNA binding exposes a universally conserved uridine at position 33 (U33), which triggers acetyl-CoA hydrolysis. We identify a series of conserved residues that are crucial for the radical-based acetylation of U34 and profile the molecular effects of patient-derived mutations. Together, we provide the high-resolution view of human Elongator and reveal its detailed mechanism of action.
Subject(s)
Cryoelectron Microscopy , RNA, Transfer , Humans , RNA, Transfer/metabolism , RNA, Transfer/chemistry , RNA, Transfer/genetics , Uridine/chemistry , Uridine/metabolism , Mutation , Acetyl Coenzyme A/metabolism , Acetyl Coenzyme A/chemistry , Models, Molecular , Acetylation , Histone Acetyltransferases/metabolism , Histone Acetyltransferases/chemistry , Histone Acetyltransferases/genetics , Protein BindingABSTRACT
BACKGROUND: Nigrostriatal microstructural integrity has been suggested as a biomarker for levodopa response in Parkinson's disease (PD), which is a strong predictor for motor response to deep brain stimulation (DBS) of the subthalamic nucleus (STN). This study aimed to explore the impact of microstructural integrity of the substantia nigra (SN), STN, and putamen on motor response to STN-DBS using diffusion microstructure imaging. METHODS: Data was collected from 23 PD patients (mean age 63 ± 7, 6 females) who underwent STN-DBS, had preoperative 3 T diffusion magnetic resonance imaging including multishell diffusion-weighted MRI with b-values of 1000 and 2000 s/mm2 and records of motor improvement available. RESULTS: The association between a poorer DBS-response and increased free interstitial fluid showed notable effect sizes (rho > |0.4|) in SN and STN, but not in putamen. However, this did not reach significance after Bonferroni correction and controlling for sex and age. CONCLUSION: Microstructural integrity of SN and STN are potential biomarkers for the prediction of therapy efficacy following STN-DBS, but further studies are required to confirm these associations.
Subject(s)
Deep Brain Stimulation , Parkinson Disease , Substantia Nigra , Subthalamic Nucleus , Humans , Deep Brain Stimulation/methods , Subthalamic Nucleus/diagnostic imaging , Subthalamic Nucleus/pathology , Female , Male , Parkinson Disease/therapy , Parkinson Disease/diagnostic imaging , Parkinson Disease/pathology , Middle Aged , Substantia Nigra/diagnostic imaging , Substantia Nigra/pathology , Aged , Diffusion Magnetic Resonance Imaging/methods , Treatment OutcomeABSTRACT
PURPOSE: Artifacts caused by metallic implants remain a challenge in computed tomography (CT). We investigated the impact of photon-counting detector computed tomography (PCD-CT) for artifact reduction in patients with orthopedic implants with respect to image quality and diagnostic confidence using different artifact reduction approaches. MATERIAL AND METHODS: In this prospective study, consecutive patients with orthopedic implants underwent PCD-CT imaging of the implant area. Four series were reconstructed for each patient (clinical standard reconstruction [PCD-CTStd], monoenergetic images at 140 keV [PCD-CT140keV], iterative metal artifact reduction (iMAR) corrected [PCD-CTiMAR], combination of iMAR and 140 keV monoenergetic [PCD-CT140keV+iMAR]). Subsequently, three radiologists evaluated the reconstructions in a random and blinded manner for image quality, artifact severity, anatomy delineation (adjacent and distant), and diagnostic confidence using a 5-point Likert scale (5 = excellent). In addition, the coefficient of variation [CV] and the relative quantitative artifact reduction potential were obtained as objective measures. RESULTS: We enrolled 39 patients with a mean age of 67.3 ± 13.2 years (51%; n = 20 male) and a mean BMI of 26.1 ± 4 kg/m2. All image quality measures and diagnostic confidence were significantly higher for the iMAR vs. non-iMAR reconstructions (all p < 0.001). No significant effect of the different artifact reduction approaches on CV was observed (p = 0.26). The quantitative analysis indicated the most effective artifact reduction for the iMAR reconstructions, which was higher than PCD-CT140keV (p < 0.001). CONCLUSION: PCD-CT allows for effective metal artifact reduction in patients with orthopedic implants, resulting in superior image quality and diagnostic confidence with the potential to improve patient management and clinical decision making.
Subject(s)
Artifacts , Metals , Tomography, X-Ray Computed , Humans , Male , Female , Aged , Prospective Studies , Tomography, X-Ray Computed/methods , Middle Aged , Prostheses and Implants , Aged, 80 and over , Photons , Radiographic Image Interpretation, Computer-Assisted/methodsABSTRACT
PURPOSE: Body composition (BC) may play a role in outcome prognostication in patients with gastroesophageal adenocarcinoma (GEAC). Artificial intelligence provides new possibilities to opportunistically quantify BC from computed tomography (CT) scans. We developed a deep learning (DL) model for fully automatic BC quantification on routine staging CTs and determined its prognostic role in a clinical cohort of patients with GEAC. MATERIALS AND METHODS: We developed and tested a DL model to quantify BC measures defined as subcutaneous and visceral adipose tissue (VAT) and skeletal muscle on routine CT and investigated their prognostic value in a cohort of patients with GEAC using baseline, 3-6-month, and 6-12-month postoperative CTs. Primary outcome was all-cause mortality, and secondary outcome was disease-free survival (DFS). Cox regression assessed the association between (1) BC at baseline and mortality and (2) the decrease in BC between baseline and follow-up scans and mortality/DFS. RESULTS: Model performance was high with Dice coefficients ≥0.94 ± 0.06. Among 299 patients with GEAC (age 63.0 ± 10.7 years; 19.4% female), 140 deaths (47%) occurred over a median follow-up of 31.3 months. At baseline, no BC measure was associated with DFS. Only a substantial decrease in VAT >70% after a 6- to 12-month follow-up was associated with mortality (hazard ratio [HR], 1.99 [95% CI, 1.18 to 3.34]; P = .009) and DFS (HR, 1.73 [95% CI, 1.01 to 2.95]; P = .045) independent of age, sex, BMI, Union for International Cancer Control stage, histologic grading, resection status, neoadjuvant therapy, and time between surgery and follow-up CT. CONCLUSION: DL enables opportunistic estimation of BC from routine staging CT to quantify prognostic information. In patients with GEAC, only a substantial decrease of VAT 6-12 months postsurgery was an independent predictor for DFS beyond traditional risk factors, which may help to identify individuals at high risk who go otherwise unnoticed.
Subject(s)
Adenocarcinoma , Deep Learning , Humans , Female , Middle Aged , Aged , Male , Artificial Intelligence , Prognosis , Adenocarcinoma/diagnostic imaging , Adenocarcinoma/surgery , Body CompositionABSTRACT
PURPOSE: CT perfusion of the brain is a powerful tool in stroke imaging, though the radiation dose is rather high. Several strategies for dose reduction have been proposed, including increasing the intervals between the dynamic scans. We determined the impact of temporal resolution on perfusion metrics, therapy decision, and radiation dose reduction in brain CT perfusion from a large dataset of patients with suspected stroke. METHODS: We retrospectively included 3555 perfusion scans from our clinical routine dataset. All cases were processed using the perfusion software VEOcore with a standard sampling of 1.5 s, as well as simulated reduced temporal resolution of 3.0, 4.5, and 6.0 s by leaving out respective time points. The resulting perfusion maps and calculated volumes of infarct core and mismatch were compared quantitatively. Finally, hypothetical decisions for mechanical thrombectomy following the DEFUSE-3 criteria were compared. RESULTS: The agreement between calculated volumes for core (ICC = 0.99, 0.99, and 0.98) and hypoperfusion (ICC = 0.99, 0.99, and 0.97) was excellent for all temporal sampling schemes. Of the 1226 cases with vascular occlusion, 14 (1%) for 3.0 s sampling, 23 (2%) for 4.5 s sampling, and 63 (5%) for 6.0 s sampling would have been treated differently if the DEFUSE-3 criteria had been applied. Reduction of temporal resolution to 3.0 s, 4.5 s, and 6.0 s reduced the radiation dose by a factor of 2, 3, or 4. CONCLUSION: Reducing the temporal sampling of brain perfusion CT has only a minor impact on image quality and treatment decision, but significantly reduces the radiation dose to that of standard non-contrast CT.
Subject(s)
Brain Ischemia , Stroke , Humans , Retrospective Studies , Drug Tapering , Stroke/diagnostic imaging , Stroke/therapy , Brain/diagnostic imaging , Brain/blood supply , Tomography, X-Ray Computed/methods , Brain Ischemia/therapy , Perfusion , Perfusion Imaging/methodsABSTRACT
PURPOSE: To investigate the value of photon-counting detector CT (PCD-CT) derived virtual non-contrast (VNC) reconstructions to identify renal cysts in comparison with conventional dual-energy integrating detector (DE EID) CT-derived VNC reconstructions. MATERIAL AND METHODS: We prospectively enrolled consecutive patients with simple renal cysts (Bosniak classification-Version 2019, density ≤ 20 HU and/or enhancement ≤ 20 HU) who underwent multiphase (non-contrast, arterial, portal venous phase) PCD-CT and for whom non-contrast and portal venous phase DE EID-CT was available. Subsequently, VNC reconstructions were calculated for all contrast phases and density as well as contrast enhancement within the cysts were measured and compared. MRI and/or ultrasound served as reference standards for lesion classification. RESULTS: 19 patients (1 cyst per patient; age 69.5 ± 10.7 years; 17 [89.5%] male) were included. Density measurements on PCD-CT non-contrast and VNC reconstructions (arterial and portal venous phase) revealed no significant effect on HU values (p = 0.301). In contrast, a significant difference between non-contrast vs. VNC images was found for DE EID-CT (p = 0.02). For PCD-CT, enhancement for VNC reconstructions was < 20 HU for all evaluated cysts. DE EID-CT measurements revealed an enhancement of > 20 HU in five lesions (26.3%) using the VNC reconstructions, which was not seen with the non-contrast images. CONCLUSION: PCD-CT-derived VNC images allow for reliable and accurate characterization of simple cystic renal lesions similar to non-contrast scans whereas VNC images calculated from DE EID-CT resulted in substantial false characterization. Thus, PCD-CT-derived VNC images may substitute for non-contrast images and reduce radiation dose and follow-up imaging.
Subject(s)
Kidney Diseases, Cystic , Tomography, X-Ray Computed , Humans , Male , Female , Aged , Prospective Studies , Tomography, X-Ray Computed/methods , Kidney Diseases, Cystic/diagnostic imaging , Middle Aged , Photons , Aged, 80 and over , Radiography, Dual-Energy Scanned Projection/methodsABSTRACT
BACKGROUND: The growing prevalence of musculoskeletal diseases increases radiologic workload, highlighting the need for optimized workflow management and automated metadata classification systems. We developed a large-scale, well-characterized dataset of musculoskeletal radiographs and trained deep learning neural networks to classify radiographic projection and body side. METHODS: In this IRB-approved retrospective single-center study, a dataset of musculoskeletal radiographs from 2011 to 2019 was retrieved and manually labeled for one of 45 possible radiographic projections and the depicted body side. Two classification networks were trained for the respective tasks using the Xception architecture with a custom network top and pretrained weights. Performance was evaluated on a hold-out test sample, and gradient-weighted class activation mapping (Grad-CAM) heatmaps were computed to visualize the influential image regions for network predictions. RESULTS: A total of 13,098 studies comprising 23,663 radiographs were included with a patient-level dataset split, resulting in 19,183 training, 2,145 validation, and 2,335 test images. Focusing on paired body regions, training for side detection included 16,319 radiographs (13,284 training, 1,443 validation, and 1,592 test images). The models achieved an overall accuracy of 0.975 for projection and 0.976 for body-side classification on the respective hold-out test sample. Errors were primarily observed in projections with seamless anatomical transitions or non-orthograde adjustment techniques. CONCLUSIONS: The deep learning neural networks demonstrated excellent performance in classifying radiographic projection and body side across a wide range of musculoskeletal radiographs. These networks have the potential to serve as presorting algorithms, optimizing radiologic workflow and enhancing patient care. RELEVANCE STATEMENT: The developed networks excel at classifying musculoskeletal radiographs, providing valuable tools for research data extraction, standardized image sorting, and minimizing misclassifications in artificial intelligence systems, ultimately enhancing radiology workflow efficiency and patient care. KEY POINTS: ⢠A large-scale, well-characterized dataset was developed, covering a broad spectrum of musculoskeletal radiographs. ⢠Deep learning neural networks achieved high accuracy in classifying radiographic projection and body side. ⢠Grad-CAM heatmaps provided insight into network decisions, contributing to their interpretability and trustworthiness. ⢠The trained models can help optimize radiologic workflow and manage large amounts of data.
Subject(s)
Deep Learning , Radiology , Humans , Artificial Intelligence , Retrospective Studies , RadiographyABSTRACT
BACKGROUND: Thalamic deep brain stimulation (DBS) is an efficacious treatment for drug-resistant essential tremor (ET) and the dentato-rubro-thalamic tract (DRT) constitutes an important target structure. However, up to 40% of patients habituate and lose treatment efficacy over time, frequently accompanied by a stimulation-induced cerebellar syndrome. The phenomenon termed delayed therapy escape (DTE) is insufficiently understood. Our previous work showed that DTE clinically is pronounced on the non-dominant side and suggested that differential involvement of crossed versus uncrossed DRT (DRTx/DRTu) might play a role in DTE development. METHODS: We retrospectively enrolled right-handed patients under bilateral thalamic DBS >12 months for ET from a cross-sectional study. They were characterized with the Fahn-Tolosa-Marin Tremor Rating Scale (FTMTRS) and Scale for the Assessment and Rating of Ataxia (SARA) scores at different timepoints. Normative fiber tractographic evaluations of crossed and uncrossed cerebellothalamic pathways and volume of activated tissue (VAT) studies together with [18F]Fluorodeoxyglucose positron emission tomography were applied. RESULTS: A total of 29 patients met the inclusion criteria. Favoring DRTu over DRTx in the non-dominant VAT was associated with DTE (R2 = 0.4463, p < 0.01) and ataxia (R2 = 0.2319, p < 0.01). Moreover, increasing VAT size on the right (non-dominant) side was associated at trend level with more asymmetric glucose metabolism shifting towards the right (dominant) dentate nucleus. CONCLUSION: Our results suggest that a disbalanced recruitment of DRTu in the non-dominant VAT induces detrimental stimulation effects on the dominant cerebellar outflow (together with contralateral stimulation) leading to DTE and thus hampering the overall treatment efficacy.
Subject(s)
Deep Brain Stimulation , Essential Tremor , Humans , Essential Tremor/diagnostic imaging , Essential Tremor/therapy , Deep Brain Stimulation/methods , Cross-Sectional Studies , Retrospective Studies , Diffusion Tensor Imaging/methods , Thalamus/diagnostic imaging , Thalamus/physiology , Treatment Outcome , AtaxiaABSTRACT
PURPOSE: Large language models (LLMs) such as ChatGPT have shown significant potential in radiology. Their effectiveness often depends on prompt engineering, which optimizes the interaction with the chatbot for accurate results. Here, we highlight the critical role of prompt engineering in tailoring the LLMs' responses to specific medical tasks. MATERIALS AND METHODS: Using a clinical case, we elucidate different prompting strategies to adapt the LLM ChatGPT using GPT4 to new tasks without additional training of the base model. These approaches range from precision prompts to advanced in-context methods such as few-shot and zero-shot learning. Additionally, the significance of embeddings, which serve as a data representation technique, is discussed. RESULTS: Prompt engineering substantially improved and focused the chatbot's output. Moreover, embedding of specialized knowledge allows for more transparent insight into the model's decision-making and thus enhances trust. CONCLUSION: Despite certain challenges, prompt engineering plays a pivotal role in harnessing the potential of LLMs for specialized tasks in the medical domain, particularly radiology. As LLMs continue to evolve, techniques like few-shot learning, zero-shot learning, and embedding-based retrieval mechanisms will become indispensable in delivering tailored outputs. KEY POINTS: · Large language models might impact radiological practice and decision-masking.. · However, implementation and performance are dependent on the assigned task.. · Optimization of prompting strategies can substantially improve model performance.. · Strategies for prompt engineering range from precision prompts to zero-shot learning..
ABSTRACT
BACKGROUND AND PURPOSE: The established global threshold of rCBF <30% for infarct core segmentation can lead to false-positives, as it does not account for the differences in blood flow between GM and WM and patient-individual factors, such as microangiopathy. To mitigate this problem, we suggest normalizing each voxel not only with a global reference value (ie, the median value of normally perfused tissue) but also with its local contralateral counterpart. MATERIALS AND METHODS: We retrospectively enrolled 2830 CTP scans with suspected ischemic stroke, of which 335 showed obvious signs of microangiopathy. In addition to the conventional, global normalization, a local normalization was performed by dividing the rCBF maps with their mirrored and smoothed counterpart, which sets each voxel value in relation to the contralateral counterpart, intrinsically accounting for GM and WM differences and symmetric patient individual microangiopathy. Maps were visually assessed and core volumes were calculated for both methods. RESULTS: Cases with obvious microangiopathy showed a strong reduction in false-positives by using local normalization (mean 14.7 mL versus mean 3.7 mL in cases with and without microangiopathy). On average, core volumes were slightly smaller, indicating an improved segmentation that was more robust against naturally low blood flow values in the deep WM. CONCLUSIONS: The proposed method of local normalization can reduce overestimation of the infarct core, especially in the deep WM and in cases with obvious microangiopathy. False-positives in CTP infarct core segmentation might lead to less-than-optimal therapy decisions when not correctly interpreted. The proposed method might help mitigate this problem.
Subject(s)
Brain Ischemia , Stroke , Humans , Brain Ischemia/therapy , Retrospective Studies , Tomography, X-Ray Computed/methods , Infarction , Cerebrovascular Circulation , Perfusion , Perfusion Imaging/methodsABSTRACT
PURPOSE: In cases of acute intracerebral hemorrhage (ICH) volume estimation is of prognostic and therapeutic value following minimally invasive surgery (MIS). The ABC/2 method is widely used, but suffers from inaccuracies and is time consuming. Supervised machine learning using convolutional neural networks (CNN), trained on large datasets, is suitable for segmentation tasks in medical imaging. Our objective was to develop a CNN based machine learning model for the segmentation of ICH and of the drain and volumetry of ICH following MIS of acute supratentorial ICH on a relatively small dataset. METHODS: Ninety two scans were assigned to training (n = 29 scans), validation (n = 4 scans) and testing (n = 59 scans) datasets. The mean age (SD) was 70 (± 13.56) years. Male patients were 36. A hierarchical, patch-based CNN for segmentation of ICH and drain was trained. Volume of ICH was calculated from the segmentation mask. RESULTS: The best performing model achieved a Dice similarity coefficient of 0.86 and 0.91 for the ICH and drain respectively. Automated ICH volumetry yielded high agreement with ground truth (Intraclass correlation coefficient = 0.94 [95% CI: 0.91, 0.97]). Average difference in the ICH volume was 1.33 mL. CONCLUSION: Using a relatively small dataset, originating from different CT-scanners and with heterogeneous voxel dimensions, we applied a patch-based CNN framework and successfully developed a machine learning model, which accurately segments the intracerebral hemorrhage (ICH) and the drains. This provides automated and accurate volumetry of the bleeding in acute ICH treated with minimally invasive surgery.
Subject(s)
Neural Networks, Computer , Tomography, X-Ray Computed , Humans , Male , Middle Aged , Aged , Aged, 80 and over , Tomography, X-Ray Computed/methods , Cerebral Hemorrhage , Machine Learning , Minimally Invasive Surgical Procedures , Image Processing, Computer-Assisted/methodsABSTRACT
OBJECTIVES: To develop a content-aware chatbot based on GPT-3.5-Turbo and GPT-4 with specialized knowledge on the German S2 Cone-Beam CT (CBCT) dental imaging guideline and to compare the performance against humans. METHODS: The LlamaIndex software library was used to integrate the guideline context into the chatbots. Based on the CBCT S2 guideline, 40 questions were posed to content-aware chatbots and early career and senior practitioners with different levels of experience served as reference. The chatbots' performance was compared in terms of recommendation accuracy and explanation quality. Chi-square test and one-tailed Wilcoxon signed rank test evaluated accuracy and explanation quality, respectively. RESULTS: The GPT-4 based chatbot provided 100% correct recommendations and superior explanation quality compared to the one based on GPT3.5-Turbo (87.5% vs. 57.5% for GPT-3.5-Turbo; P = .003). Moreover, it outperformed early career practitioners in correct answers (P = .002 and P = .032) and earned higher trust than the chatbot using GPT-3.5-Turbo (P = 0.006). CONCLUSIONS: A content-aware chatbot using GPT-4 reliably provided recommendations according to current consensus guidelines. The responses were deemed trustworthy and transparent, and therefore facilitate the integration of artificial intelligence into clinical decision-making.
Subject(s)
Artificial Intelligence , Software , Humans , Clinical Decision-Making , Cone-Beam Computed Tomography , ConsensusABSTRACT
OBJECTIVES: To determine the diagnostic accuracy of ultra-high-resolution photon-counting detector CT angiography (UHR PCD-CTA) for evaluating coronary stent patency compared to invasive coronary angiography (ICA). METHODS: Consecutive, clinically referred patients with prior coronary stent implantation were prospectively enrolled between August 2022 and March 2023 and underwent UHR PCD-CTA (collimation, 120 × 0.2 mm). Two radiologists independently analyzed image quality of the in-stent lumen using a 5-point Likert scale, ranging from 1 ("excellent") to 5 ("non-diagnostic"), and assessed all coronary stents for the presence of in-stent stenosis (≥ 50% lumen narrowing). The diagnostic accuracy of UHR PCD-CTA was determined, with ICA serving as the standard of reference. RESULTS: A total of 44 coronary stents in 18 participants (mean age, 83 years ± 6 [standard deviation]; 12 women) were included in the analysis. In 3/44 stents, both readers described image quality as non-diagnostic, whereas reader 2 noted a fourth stent to have non-diagnostic image quality. In comparison to ICA, UHR PCD-CTA demonstrated a sensitivity, specificity, and accuracy of 100% (95% CI [confidence interval] 47.8, 100), 92.3% (95% CI 79.1, 98.4), and 93.2% (95% CI 81.3, 98.6) for reader 1 and 100% (95% CI 47.8, 100), 87.2% (95% CI 72.6, 95.7), and 88.6% (95% CI 75.4, 96.2) for reader 2, respectively. Both readers observed a 100% negative predictive value (36/36 stents and 34/34 stents). Stent patency inter-reader agreement was 90.1%, corresponding to a substantial Cohen's kappa value of 0.72. CONCLUSIONS: UHR PCD-CTA enables non-invasive assessment of coronary stent patency with high image quality and diagnostic accuracy. CLINICAL RELEVANCE STATEMENT: Ultra-high-resolution photon-counting detector CT angiography represents a reliable and non-invasive method for assessing coronary stent patency. Its high negative predictive value makes it a promising alternative over invasive coronary angiography for the rule-out of in-stent stenosis. KEY POINTS: ⢠CT-based evaluation of coronary stent patency is limited by stent-induced artifacts and spatial resolution. ⢠Ultra-high-resolution photon-counting detector CT accurately evaluates coronary stent patency compared to invasive coronary angiography. ⢠Photon-counting detector CT represents a promising method for the non-invasive rule-out of in-stent stenosis.