-
Defining tumor growth in vestibular schwannomas: a volumetric inter-observer variability study in contrast-enhanced T1-weighted MRI medRxiv. Radiol. Imaging Pub Date : 2024-03-16 Stefan Cornelissen, Sammy M. Schouten, Patrick P.J.H. Langenhuizen, Suan Te Lie, Henricus P.M. Kunst, Peter H.N. de With, Jeroen B. Verheul
Objective: For patients with vestibular schwannomas (VS), a conservative observational approach is increasingly used. Therefore, the need for accurate and reliable volumetric tumor monitoring is important. Currently, a volumetric cutoff of 20% increase in tumor volume is widely used to define tumor growth in VS. The goal of this study is to investigate the tumor volume dependency on the limits of agreement
-
Normalizing Spinal Cord Compression Morphometric Measures: Application in Degenerative Cervical Myelopathy medRxiv. Radiol. Imaging Pub Date : 2024-03-15 Sandrine Bédard, Jan Valošek, Maryam Seif, Armin Curt, Simon Schading, Nikolai Pfender, Patrick Freund, Markus Hupp, Julien Cohen-Adad
Objective: Automatic and robust characterization of spinal cord shape from MRI images is relevant to assess the severity of spinal cord compression in degenerative cervical myelopathy (DCM) and to guide therapeutic strategy. Despite its popularity, the maximum spinal cord compression (MSCC) index has practical limitations to objectively assess the severity of cord compression. Firstly, it is computed
-
Cross-validation coordinate analysis (CVCA): performing coordinate based meta-analysis using cross-validation medRxiv. Radiol. Imaging Pub Date : 2024-03-15 Christopher R Tench
Coordinate based meta-analysis (CBMA) can be used to estimate where a future neuroimaging study testing a particular hypothesis might report summary results (activation foci, for example). However, current methods cannot be validated for all possible analyses, and use empirical features that might not always be ideal. Indeed, the various algorithms that perform CBMA tend to produce somewhat different
-
Dissecting unique and common variance across body and brain health indicators using age prediction medRxiv. Radiol. Imaging Pub Date : 2024-03-14 Dani Beck, Ann-Marie G. de Lange, Tiril P Gurholt, Irene Voldsbekk, Ivan I. Maximov, Sivaniya Subramaniapillai, Louise Schindler, Guy Hindley, Esten H. Leonardsen, Zillur Rahman, Dennis van der Meer, Max Korbmacher, Jennifer Linge, Olof D Leinhard, Karl T Kalleberg, Andreas Engvig, Ida Sonderby, Ole A Andreassen, Lars T Westlye
Ageing is a heterogeneous multisystem process involving different rates of decline in physiological integrity across biological systems. The current study dissects the unique and common variance across body and brain health indicators and parses inter-individual heterogeneity in the multisystem ageing process. Using machine-learning regression models on the UK Biobank dataset (N = 32,593, age range
-
Performance and change impact analysis of a commercial artificial intelligence tool for radiographic knee osteoarthritis grading and joint space width measuring medRxiv. Radiol. Imaging Pub Date : 2024-03-14 Mathias Willadsen Brejneboel, Anders Lenskjold, Jacob J. Visser, Huib Ruitenbeek, Katharina Ziegeler, Philip Hansen, Janus Uhd Nybing, Kay Geert A. Hermann, Edwin H.G. Oei, Mikael Boesen
Background and rationale: Knee osteoarthritis (OA) is a common disease characterized by reduced function, stiffness, and pain. This clinical diagnosis is commonly supported with radiography of the weight-bearing knee. Radiographic features, such as the Kellgren-Lawrence (KL) grading system, are used as eligibility criteria for clinical studies while others, such as the OARSI grades and minimal joint
-
Towards a multimodal neuroimaging-based risk score for mild cognitive impairment by combining clinical studies with a large (N>37000) population-based study medRxiv. Radiol. Imaging Pub Date : 2024-03-14 Elaheh Zendehrouh, Mohammad SE Sendi, Anees Abrol, Ishaan Batta, Reihaneh Hassanzadeh, Vince Calhoun
Alzheimer's disease (AD) is the most common form of age-related dementia, leading to a decline in memory, reasoning, and social skills. While numerous studies have investigated the genetic risk factors associated with AD, less attention has been given to identifying a brain imaging-based measure of AD risk. This study introduces a novel approach to assess mild cognitive impairment MCI, as a stage before
-
A novel use of diffusion-weighted whole-body magnetic resonance imaging with background body signal suppression to diagnose infectious aortitis medRxiv. Radiol. Imaging Pub Date : 2024-03-13 JIEN SAITO, Masahiro Muto, Masafumi Tada, Isao Yokota, Shinji Kamiya, Yukihide Numata, Hideki Sasaki, Takuya Hashizume, Miki Asano, Satoru Wakasa
Background: Diffusion-weighted whole-body imaging with background body signal suppression is one of the whole-body magnetic resonance imaging techniques and is effective in diagnosing inflammatory and infectious diseases. We aimed to evaluate the diagnostic performance of this modality in infectious aortitis, which remains unclear. Methods: The study participants were 32 patients with suspected infectious
-
LST-AI: a Deep Learning Ensemble for Accurate MS Lesion Segmentation medRxiv. Radiol. Imaging Pub Date : 2024-03-11 Tun Wiltgen, Julian McGinnis, Sarah Schlaeger, Florian Kofler, CuiCi Voon, Achim Berthele, Daria Bischl, Lioba Grundl, Nikolaus Will, Marie Metz, David Schinz, Dominik Sepp, Philipp Prucker, Benita Schmitz-Koep, Claus Zimmer, Bjoern Menze, Daniel Rueckert, Bernhard Hemmer, Jan Kirschke, Mark Muhlau, Benedikt Wiestler
Automated segmentation of brain white matter lesions is crucial for both clinical assessment and scientific research in multiple sclerosis (MS). Over a decade ago, we introduced an engineered lesion segmentation tool, LST. While recent lesion segmentation approaches have leveraged artificial intelligence (AI), they often remain proprietary and difficult to adopt. As an open-source tool, we present
-
Investigation of diffusion time dependence of apparent diffusion coefficient and intravoxel incoherent motion parameters in the human kidney medRxiv. Radiol. Imaging Pub Date : 2024-03-10 Julia Stabinska, Thomas Andreas Thiel, Helge Jörn Zöllner, Thomas Benkert, Hans-Jörg Wittsack, Alexandra Ljimani
Purpose: To characterize the diffusion time (Δeff) dependence of apparent diffusion coefficient (ADC) and intravoxel incoherent motion (IVIM)-related parameters in the human kidney at 3T. Methods: Sixteen healthy volunteers underwent an MRI examination at 3T including DWI at different Δeff ranging from 24.1 ms to 104.1 ms. The extended mono-exponential ADC, and IVIM models were fitted to the data for
-
Artificial intelligence-generated smart impression from 9.8-million radiology reports as training datasets from multiple sites and imaging modalities medRxiv. Radiol. Imaging Pub Date : 2024-03-09 Parisa Kaviani, Mannudeep K Kalra, Subba R Digumarthy, Karen Rodriguez, Sheela Agarwal, Rupert Brooks, Sovann En, Tarik Alkasab, Bernardo C Bizzo, Keith J Dreyer
Importance: Automatic generation of the impression section of radiology report can help make radiologists efficient and avoid reporting errors. Objective: To evaluate the relationship, content, and accuracy of an Powerscribe Smart Impression (PSI) against the radiologists reported findings and impression (RDF). Design, Setting, and Participants: The institutional review board approved retrospective
-
Impact of Multimodal Prompt Elements on Diagnostic Performance of GPT-4(V) in Challenging Brain MRI Cases medRxiv. Radiol. Imaging Pub Date : 2024-03-06 Severin Schramm, Silas Preis, Marie-Christin Metz, Kirsten Jung, Benita Schmitz-Koep, Claus Zimmer, Benedikt Wiestler, Dennis Martin Hedderich, Su Hwan Kim
Background Recent studies have explored the application of multimodal large language models (LLMs) in radiological differential diagnosis. Yet, how different multimodal input combinations affect diagnostic performance is not well understood. Purpose To evaluate the impact of varying multimodal input elements on the accuracy of GPT-4(V)-based brain MRI differential diagnosis. Methods Thirty brain MRI
-
An experimental evaluation of the relationship between the induced radiofrequency heating near an implanted conductive medical device during MRI, scanner reported B1+rms, and scanner reported average transmit power medRxiv. Radiol. Imaging Pub Date : 2024-03-06 David H Gultekin, J. Thomas Vaughan, Devashish Shrivastava
Background: Time-varying radiofrequency (RF) fields necessary to perform magnetic resonance imaging (MRI) may induce excessive heating near implanted conductive medical devices during MRI. Both time and space-averaged root mean square of the effective magnetic field (B1+rms) and whole-body average specific absorption rate (SAR) (average RF power per unit body weight) have been proposed as metrics to
-
High Resolution Multi-delay Arterial Spin Labeling with Transformer based Denoising for Pediatric Perfusion MRI medRxiv. Radiol. Imaging Pub Date : 2024-03-06 Qinyang Shou, Chenyang Zhao, Xingfeng Shao, Megan Herting, Danny JJ Wang
Multi-delay arterial spin labeling (MDASL) can quantitatively measure cerebral blood flow (CBF) and arterial transit time (ATT), which is particularly suitable for pediatric perfusion imaging. Here we present a high resolution (iso-2mm) MDASL protocol and performed test-retest scans on 21 typically developing children aged 8 to 17 years. We further proposed a Transformer-based deep learning (DL) model
-
Automated segmentation of brain metastases in T1-weighted contrast-enhanced MR images pre and post Stereotactic Radiosurgery medRxiv. Radiol. Imaging Pub Date : 2024-03-05 Hemalatha Kanakarajan, Wouter De Baene, Patrick Hanssens, Margriet Sitskoorn
Background and purpose: Accurate segmentation of brain metastases on Magnetic Resonance Imaging (MRI) is a tedious and time-consuming task for radiologists that could be optimized with deep learning (DL) methods. Previous studies that evaluated the performance of several DL algorithms focused on training and testing the models on the planning MRI only. The purpose of this study is to evaluate well-known
-
Radiomic Profiling of Lung CT in a Cohort of Sarcoidosis Cases medRxiv. Radiol. Imaging Pub Date : 2024-03-05 Nichole E Carlson, William Lippitt, Zac Buchalski, Sarah M Ryan, Margaret Mroz, Brianna Barkes, Shu-Yi Liao, Lisa A Meier, Tasha E Fingerlin
Background: High resolution computed tomography (HRCT) of the chest is increasingly used in clinical practice for sarcoidosis. Visual assessment of chest HRCTs in patients with sarcoidosis has high inter- and intra-rater variation. Radiomics offers a reproducible quantitative assessment of HRCT lung parenchyma and could be useful as an additional summary measure of disease. We develop radiomic profiles
-
Flexible-circuit-based 3-D aware modular optical brain imaging system for high-density measurements in natural settings medRxiv. Radiol. Imaging Pub Date : 2024-03-04 Edward Xu, Morris Vanegas, Miguel Adrian Mireles Nunez, Artem Dementyev, Meryem Yucel, Stefan Carp, Qianqian Fang
Significance: Functional near-infrared spectroscopy (fNIRS) presents an opportunity to study human brains in everyday activities and environments. However, achieving robust measurements under such dynamic condition remains a significant challenge. Aim: The modular optical brain imaging (MOBI) system is designed to enhance optode-to-scalp coupling and provide real-time probe 3-D shape estimation to
-
Disorder-Free Data are All You Need: Inverse Supervised Learning for Broad-Spectrum Head Disorder Detection medRxiv. Radiol. Imaging Pub Date : 2024-03-03 Yuwei He, Yuchen Guo, Jinhao Lv, Liangd Ma, Haotian Tan, Wei Zhang, Guiguang Ding, Hengrui Liang, Jianxing He, Xi Lou, Qiongha Dai, Feng Xu
Collecting and annotating sufficient data containing disorders is crucial for the development of artificial intelligence (AI)-based medical systems. However, preparing data with complete disorder types and adequate annotations is challenging, which limits the ability of existing AI-based medical systems to diagnose specific disorders. In this paper, we introduce a novel AI-based system that achieves
-
The impact of localization and registration accuracy on estimates of deep brain stimulation electrode position in stereotactic space medRxiv. Radiol. Imaging Pub Date : 2024-03-02 Mohamad Abbass, Greydon Gilmore, Brendan Santyr, Alan Chalil, Alaa Taha, Mandar Jog, Keith MacDougall, Andrew G. Parrent, Terry M. Peters, Jonathan C. Lau
Effects of deep brain stimulation (DBS) depend on millimetric accuracy and are commonly studied across populations by registering patient scans to a stereotactic space. Multiple factors contribute to estimates of electrode position, but the millimetric contributions of these factors remains poorly quantified. We previously validated 32 anatomical fiducials (AFIDs) to measure AFID registration error
-
PACT-3D, a Deep Learning Algorithm for Pneumoperitoneum Detection in Abdominal CT Scans medRxiv. Radiol. Imaging Pub Date : 2024-03-02 I-Min Chiu, Teng-Yi Huang, Kuei-Hong Kuo
Pneumoperitoneum, necessitates surgical intervention in 85-90% of cases, relies heavily on CT scans for diagnosis. Delay or misdiagnosis in detecting pneumoperitoneum can significantly increase mortality and morbidity. Our study introduced PACT-3D, a deep learning model developed to identify pneumoperitoneum in CT images. In this single hospital study, we retrospectively reviewed abdominal CT scans
-
Cortical thickness and grey-matter volume anomaly detection in individual MRI scans: Comparison of two methods medRxiv. Radiol. Imaging Pub Date : 2024-02-29 David Romascano, Michael Rebsamen, Piotr Radojewski, Timo Blattner, Richard McKinley, Roland Wiest, Christian Rummel
Over the past decades, morphometric analysis of brain MRI has contributed substantially to the understanding of healthy brain structure, development and aging as well as to improved characterisation of disease related pathologies. Certified commercial tools based on normative modeling of these metrics are meanwhile available for diagnostic purposes, but they are cost intensive and their clinical evaluation
-
Cerebrovascular reactivity dispersion as a new biomarker of recent stroke symptomatology in moyamoya medRxiv. Radiol. Imaging Pub Date : 2024-02-29 Caleb J Han, Wesely T Richerson, Maria Garza, Mark Rodeghier, Murli Mishra, Larry Taylor Davis, Matthew Fusco, Rohan V Chitale, Shuhei Shiino, Lori C. Jordan, Manus Joseph Donahue
Background: Moyamoya disease (MMD) is a non-atherosclerotic intracranial steno-occlusive condition placing patients at high risk for ischemic stroke. Direct and indirect surgical revascularization can improve blood flow in MMD; however, randomized trials demonstrating efficacy have not been performed and biomarkers of parenchymal hemodynamic impairment are needed to triage patients for interventions
-
Improving 3D-CINE tTV-regularized whole-heart MRI reconstruction medRxiv. Radiol. Imaging Pub Date : 2024-02-28 Bastien Milani, Christopher Roy, Jean-Baptiste Ledoux, David Rotzinger, Ambra Masi, Renaud Troxler, Salim Si-Mohamed, Jerome Yerly, Ludovica Romanin, Tobias Rutz, Estelle Tenisch, Milan Prsa, Juerg Schwitter, Matthias Stuber
Purpose: To improve the image quality of 3D radial free-running MRI data of the heart through a deliberate and stepwise extension of the XD-GRASP reconstruction. Methods: Ferumoxytol-enhanced cardiac free-running 3D-radial data were reconstructed using an XD-GRASP reconstruction improved by 4 new developments: motion-compensated temporal-Total-Variation (MC–tTV) regularization for 3D images, a new
-
seg-metrics: a Python package to compute segmentation metrics medRxiv. Radiol. Imaging Pub Date : 2024-02-23 Jingnan Jia, Marius Staring, Berend C Stoel
Medical image segmentation (MIS) is an important task in medical image process- ing. Unfortunately, there is not a out-of-the-box python package for the evaluation metrics of MIS. Therefore, we developed seg-metrics, an open-source Python package for MIS model evaluation. Unlike existing packages, seg-metrics offers user-friendly interfaces for various overlap-based and distance-based metrics, providing
-
Constructing personalized characterizations of structural brain aberrations in patients with dementia and mild cognitive impairment using explainable artificial intelligence medRxiv. Radiol. Imaging Pub Date : 2024-02-22 Esten H Leonardsen, Karin Persson, Edvard Grødem, Nicola Dinsdale, Till Schellhorn, James M Roe, Didac Vidal-Piñeiro, Øystein Sørensen, Tobias Kaufmann, Eric Westman, Andre Marquand, Geir Selbæk, Ole A Andreassen, Thomas Wolfers, Lars T Westlye, Yunpeng Wang
Deep learning approaches for clinical predictions based on magnetic resonance imaging data have shown great promise as a translational technology for diagnosis and prognosis in neurological disorders, but its clinical impact has been limited. This is partially attributed to the opaqueness of deep learning models, causing insufficient understanding of what underlies their decisions. To overcome this
-
Transcranial Blood-Brain Barrier Opening in Alzheimer's Disease Patients Using A Portable Focused Ultrasound System with Real-Time 2-D Cavitation Mapping medRxiv. Radiol. Imaging Pub Date : 2024-02-21 Sua Bae, Keyu Liu, Antonios N Pouliopoulos, Robin Ji, Sergio Jimenez-Gambin, Omid Yousefian, Alina R Kline-Schoder, Alec Batts, Danae Kokossis, Akiva Mintz, Lawrence S Honig, Elisa E Konofagou
Background: Focused ultrasound (FUS) in combination with microbubbles has recently shown great promise in facilitating blood-brain barrier (BBB) opening for drug delivery and immunotherapy in Alzheimer's disease (AD). However, it is currently limited to systems integrated within the MRI suites or requiring post-surgical implants, thus restricting its widespread clinical adoption. In this pilot study
-
Electrode positioning errors reduce current dose for focal tDCS set-ups: Evidence from individualized electric field mapping medRxiv. Radiol. Imaging Pub Date : 2024-02-20 Filip Niemann, Steffen Riemann, Ann-Kathrin Hubert, Daria Antonenko, Axel Thielscher, Andrew K Martin, Nina Unger, Agnes Flöel, Marcus F Meinzer
Objective: Electrode positioning errors contribute to variability of transcranial direct current stimulation (tDCS) effects. We investigated the impact of electrode positioning errors on current flow for tDCS set-ups with different focality. Methods: Deviations from planned electrode positions were determined using data acquired in an experimental study (N=240 datasets) that administered conventional
-
A computationally efficient method for dimensionality reduction in multi-channel spectral CT medRxiv. Radiol. Imaging Pub Date : 2024-02-20 Olivia F Sandvold, Roland Proksa, Heiner Daerr, Amy E Perkins, Kevin M Brown, Thomas Koehler, Ravindra M Manjeshwar, Peter B Noel
Objective Multi-channel spectral CT technology holds the promise of measuring incoming X-rays across various energy spectrums, significantly enhancing iodine imaging performance. Nonetheless, it necessitates the reduction of data dimensions, such as condensing multiple energy bins, to ensure processing times remain within clinically acceptable limits and to maintain compatibility with traditional two-dimensional
-
Predicting gestational age at birth in the context of preterm birth from multi-modal fetal MRI medRxiv. Radiol. Imaging Pub Date : 2024-02-18 Diego Fajardo-Rojas, Megan Hall, Daniel Cromb, Mary A. Rutherford, Lisa Story, Emma Robinson, Jana Hutter
Preterm birth is associated with significant mortality and a risk for lifelong morbidity. The complex multifactorial aetiology hampers accurate prediction and thus optimal care. A pipeline consisting of bespoke machine learning methods for data imputation, feature selection, and regression models to predict gestational age (GA) at birth was developed and evaluated from comprehensive multi-modal morphological
-
Enhancing Semantic Segmentation in Chest X-Ray Images through Image Preprocessing: ps-KDE for Pixel-wise Substitution by Kernel Density Estimation medRxiv. Radiol. Imaging Pub Date : 2024-02-17 Yuanchen Wang, Yujie Guo, Ziqi Wang, Linzi Yu, Yujie Yan, Zifan Gu
Background: Deep-learning-based semantic segmentation algorithms, in combination with image preprocessing techniques, can reduce the need for human annotation and advance disease classification. Among established preprocessing techniques, CLAHE has demonstrated efficacy in enhancing the segmentations algorithms across various modalities. Method: This study proposes a novel preprocessing technique,
-
Probing the Depths for Diagnostic Performance of Biparametric and Multiparametric MRI for Prostate Cancer Detection- A Meta-Analysis medRxiv. Radiol. Imaging Pub Date : 2024-02-14 Dev Desai, Vismit Gami, Abhijay Shah, Dwija Raval, Parth Gupta, Hetvi Shah
Background- Prostate cancer is a malignancy that originates in the prostate gland,and can vary in aggressiveness, often requiring a combination of diagnostic methods and imaging for accurate detection and management. In an attempt to ensure timely diagnosis and prevent complications, the choice of the right diagnostic modality becomes crucial. MRI has gained prominence in the diagnosis of prostate
-
Pulsatility analysis of the circle of Willis medRxiv. Radiol. Imaging Pub Date : 2024-02-14 Henning U. Voss, Qolamreza R. Razlighi
Purpose: To evaluate the phenomenological significance of cerebral blood pulsatility imaging in aging research. Methods: N = 38 subjects aged from 20 to 72 years of age (24 females) were imaged with ultrafast MRI with a sampling rate of 100 ms and simultaneous acquisition of pulse oximetry data. Of these, 28 subjects had acceptable MRI and pulse data, with 16 subjects between 20 and 28 years of age
-
Quantification of [11C]ABP688 binding in human brain using cerebellum as reference region: biological interpretation and limitations medRxiv. Radiol. Imaging Pub Date : 2024-02-13 Michele Stanislaw Milella, luciano minuzzi, chawki benkelfat, jean-paul soucy, alexandre kirlow, esther Schirrmacher, mark angle, Jeroen Verhaeghe, gassan massarweh, andrew j reader, antonio aliaga, Jose Eduardo Peixoto-Santos, Marie-Christine guiot, eliane kobayashi, pedro rosa-neto, Marco Leyton
In vitro data from primates provide conflicting evidence about the cerebellum suitability as a reference region for quantifying type 5 metabotropic glutamate receptor (mGluR5) binding parameters with positron emission tomography (PET). To address this, we first measured mGluR5 density in postmortem human cerebellum using [3H]ABP688 autoradiography (n=5) and immunohistochemistry (n=6). Next, in vivo
-
Liver Intrinsic Function Evaluation (LIFE): Multi-parametric Liver Function Profiles of Patients Undergoing Hepatectomy medRxiv. Radiol. Imaging Pub Date : 2024-02-13 Christian Simonsson, Wolf Claus Bartholoma, Anna Lindhoff-Larsson, Markus Karlsson, Shan Cai, Jens Tellman, Bengt Noren, Bergthor Bjornsson, Gunnar Cedersund, Nils Dahlstrom, Per Sandstrom, Peter Lundberg
For a range of liver malignancies, the only curative treatment option may be hepatectomy, which may have fatal complications. Therefore, an unbiased pre-operative risk assessment is vital, however, at present the assessment is typically based on global liver function only. Magnetic resonance imaging (MRI) modalities have the possibility to aid this assessment, by introducing additional characterization
-
Conformal Triage for Medical Imaging AI Deployment medRxiv. Radiol. Imaging Pub Date : 2024-02-11 Anastasios Nikolas Angelopoulos, Stuart R Pomerantz, Synho Do, Stephen Bates, Christopher P Bridge, Daniel C Elton, Michael H Lev, R Gilberto Gonzalez, Michael I Jordan, Jitendra Malik
Background: The deployment of black-box AI models in medical imaging presents significant challenges, especially in maintaining reliability across different clinical settings. These challenges are compounded by distribution shifts that can lead to failures in reproducing the accuracy attained during the AI model's original validations. Method: We introduce the conformal triage algorithm, designed to
-
Improving Adherence to Pulmonary Embolism Evaluation Consensus Guidelines: A Multimodal Approach medRxiv. Radiol. Imaging Pub Date : 2024-02-11 Ethan D'Silva, Syed Zaidi, Phil Ramis, Eric M Rohren
Objective: Despite the publication of multiple widely accepted guidelines, Computed Tomography pulmonary angiography (CTPA) remains overutilized in the emergency (ED) setting for evaluating suspected pulmonary embolism (PE). We developed and evaluated a multimodal program that aimed to improve ordering provider adherence to CTPA use guidelines, and consequently positivity rate. Methods: We retrospectively
-
Brain 3T magnetic resonance imaging in neonates: features and incidental findings from a research cohort enriched for preterm birth medRxiv. Radiol. Imaging Pub Date : 2024-02-08 Gemma Sullivan, Alan J. Quigley, Samantha Choi, Rory Teed, Manuel Blesa Cabez, Kadi Vaher, Amy E. Corrigan, David Q. Stoye, Michael J. Thrippleton, Mark E. Bastin, James P. Boardman
Background and objectives The survival rate and patterns of brain injury after very preterm birth are evolving with changes in clinical practices. Additionally, incidental findings can present legal, ethical and practical considerations. Here, we report MRI features and incidental findings from a large, contemporary research cohort of very preterm infants and term controls.
-
Percutaneous Nephrostomy Guidance by a Convolutional Neural Network Based Endoscopic Optical Coherence Tomography System medRxiv. Radiol. Imaging Pub Date : 2024-02-07 Chen Wang, Paul Calle, Feng Yan, Qinghao Zhang, Kar-ming Fung, Chongle Pan, Qinggong Tang
Percutaneous nephrostomy (PCN) is a commonly used procedure for kidney surgeries. However, difficulties persist in precisely locating the PCN needle tip during its insertion into the kidney. Challenges for PCN needle guidance exist in two aspects: 1) Accurate tissue recognition, and 2) Renal blood vessel detection. In this study, we demonstrated an endoscopic optical coherence tomography (OCT) system
-
Self-supervised Learning for Chest CT - Training Strategies and Effect on Downstream Applications medRxiv. Radiol. Imaging Pub Date : 2024-02-05 Amara Tariq, Bhavik Patel, Imon Banerjee
Self-supervised pretraining can reduce the amount of labeled training data needed by pre-learning fundamental visual characteristics of the medical imaging data. In this study, we investigate several self-supervised training strategies for chest computed tomography exams and their effects of downstream applications. we benchmark five well-known self-supervision strategies (masked image region prediction
-
Patient Centric Summarization of Radiology Findings using Large Language Models medRxiv. Radiol. Imaging Pub Date : 2024-02-05 Amara Tariq, Aisha Urooj, Shubham Trivedi, Sam Fathizadeh, Gokul Ramasamy, Nelly Tan, Matthew Stib, Bhavik Patel, Imon Banerjee
Objective. Develop automated AI models for patient-sensitive summarization of radiology reports. Level of medical education or socio-economic background of a patient may dictate their level of understanding of medical jargon. Inability to understand primary findings from a radiology report may lead to unnecessary anxiety among patients or result in missed follow up. Materials and Methods. Computed
-
Human-AI Collaboration in Large Language Model-Assisted Brain MRI Differential Diagnosis: A Usability Study medRxiv. Radiol. Imaging Pub Date : 2024-02-06 Su Hwan Kim, Severin Schramm, Cornelius Berberich, Enrike Rosenkranz, Lena Schmitzer, Kerem Serguen, Christopher Klenk, Nicolas Lenhart, Claus Zimmer, Benedikt Wiestler, Dennis M. Hedderich
Background Prior studies have shown the potential of large language models (LLMs) to support in differential diagnosis in radiology. However, the interaction of human users with LLMs in this context has not been evaluated.
-
A Nomogram to Predict Pneumothorax Requiring Chest Tube Placement following Percutaneous CT-guided Lung Biopsy medRxiv. Radiol. Imaging Pub Date : 2024-02-03 Masha Bondarenko, Jianxiang Zhang, Ulysis Baal, Brian Lam, Gunvant R Chaudhari, Yoo Jin Lee, Jamie Schroeder, Maya Vella, Brian Haas, Thienkhai Huy Vu, Jonathan Liu, Kimberly Kallianos, Shravan Sridhar, Brett Elicker, Jae Ho Sohn
Background: Pneumothorax requiring chest tube after CT-guided transthoracic lung biopsy is one of the common complications, and the required hospital stay after chest tube placement represents an added clinical risk to patients and cost to the healthcare system. Identifying high-risk patients can prompt alternative biopsy modes and/or better preparation for more focused post-procedural care. Purpose:
-
First positronium image of the human brain in vivo medRxiv. Radiol. Imaging Pub Date : 2024-02-03 Pawel Moskal, Jakub Baran, Steven Bass, Jaroslaw Choinski, Neha Chug, Catalina Curceanu, Eryk Czerwinski, Meysam Dadgar, Manish Das, Kamil Ł Dulski, Kavya Valsan Eliyan, Katarzyna Fronczewska, Aleksander Gajos, Krzysztof Kacprzak, Marcin Kajetanowicz, Tevfik Kaplanoglu, Lukasz Kaplon, Konrad Klimaszewski, Malgorzata Kobylecka, Grzegorz Korcyl, Tomasz Kozik, Wojciech Krzemien, Karol Kubat, Deepak Kumar
Positronium, an unstable atom consisting of an electron and a positron, is abundantly produced within the molecular voids of a patient's body during positron emission tomography (PET) diagnosis. Its properties, such as its average lifetime between formation and annihilation into photons, dynamically respond to the submolecular architecture of the tissue and the partial pressure of oxygen molecules
-
AutoCumulus: an Automated Mammographic Density Measure Created Using Artificial Intelligence medRxiv. Radiol. Imaging Pub Date : 2024-02-03 Osamah Al-qershi, Tuong L Nguyen, Michael E Elliott, Daniel F Schmidt, Enes Makalic, Shuai Li, Samantha K Fox, James Dowty, Carlos A Pena-Solorzano, Chun Fung Kwok, Yuanhong Chen, Chong Wang, Jocelyn Lippey, Peter Brotchie, Gustavo Carneiro, Davis J McCarthy, Yeojin Jeong, Joohon Sung, Helen ML Frazer, John L Hopper
Background : Mammographic (or breast) density is an established risk factor for breast cancer. There are a variety of approaches to measurement including quantitative, semi-automated and automated approaches. We present a new automated measure, AutoCumulus, learnt from applying deep learning to semi-automated measures. Methods: We used mammograms of 9,057 population-screened women in the BRAIx study
-
Automated Deep Learning Segmentation of Cardiac Inflammatory FDG PET medRxiv. Radiol. Imaging Pub Date : 2024-02-03 Alexis Poitrasson-Rivière, Michael Vanderver, Tomoe Hagio, Liliana Arida-Moody, Jonathan B Moody, Jennifer M Renaud, Edward P Ficaro, Venkatesh L. Murthy
Background: Fluorodeoxyglucose positron emission tomography (FDG PET) with glycolytic metabolism suppression plays a pivotal role in diagnosing cardiac sarcoidosis. Reorientation of images to match perfusion datasets is critical and myocardial segmentation enables consistent image scaling and quantification. However, both are challenging and labor intensive. We developed a 3D U-Net deep learning (DL)
-
A machine-learning model to harmonize brain volumetric data for quantitative neuro-radiological assessment of Alzheimer′s disease medRxiv. Radiol. Imaging Pub Date : 2024-02-03 Damiano Archetti, Vikram Venkatraghavan, Bela Weiss, Pierrick Bourgeat, Tibor Auer, Zoltan Vidnyanszky, Stanley Durrleman, Wiesje M van der Flier, Frederik Barkhof, Daniel C Alexander, Andre Altmann, Alberto Redolfi, Betty Tijms, Neil P Oxtoby
Background: Structural MRI plays a pivotal role in the radiological workup for assessing neurodegeneration. Scanner-related differences hinder quantitative neuroradiological assessment of Alzheimer′s disease (QNAD). This study aims to train a machine-learning model to harmonize brain volumetric data of patients not encountered during model training. Method: Neuroharmony is a recently developed method
-
Resting-state Functional Connectivity Predicts Cochlear-Implant Speech Outcomes medRxiv. Radiol. Imaging Pub Date : 2024-02-02 Jamal Esmaelpoor, Tommy Peng, Beth Jelfs, Darren Mao, Maureen Shader, Colette McKay
Background: Cochlear implants (CIs) have significantly improved hearing restoration for individuals with severe or profound hearing loss. Nonetheless, a substantial variability in CI outcomes remains unexplained, even when considering subject-specific factors such as age and duration of deafness. This study investigates the use of resting-state functional near-infrared spectroscopy (fNIRS) recordings
-
RAMDS: Retrieval Augmented Medical Diagnosis System for Explainable Breast Cancer Classification from Ultrasound Images. medRxiv. Radiol. Imaging Pub Date : 2024-02-02 Johnson Thomas, Ethan Thomas Johnson, Jathin K Bande
Breast cancer, a leading cause of cancer-related deaths in women, presents a growing challenge in medical diagnostics. Despite the effectiveness of mammography and ultrasound, the ambiguity in non-invasive scans often necessitates invasive procedures. Our primary goal was to create an AI model that could predict breast cancer with high negative predictive value and reduce unnecessary procedures. This
-
Artificial-Intelligence Powered Identification of High-Risk Breast Cancer Subgroups Using Mammography : A Multicenter Study Integrating Automated Brightest Density Measures with Deep Learning Metrics medRxiv. Radiol. Imaging Pub Date : 2024-01-29 Yeojin Jeong, Jeesoo Lee, Young-jin Lee, Jiyun Hwang, Sae Byul Lee, Tae-Kyung Yoo, Myeong-Seong Kim, Jae Il Kim, John L Hopper, Tuong L Nguyen, Jong Won Lee, Joohon Sung
Mammography plays a crucial role in breast cancer (BC) risk assessment. Recent breakthroughs show that deep learning (DL) in mammography is expanding from diagnosis to effective risk prediction. Moreover, the brightest mammographic breast density (MBD), termed "cirrocumulus", signifies an authentic risk. Addressing the challenges in quantifying above recent measures, we present MIDAS: a DL-derived
-
Quantitative MRI biomarker for classification of clinically significant prostate cancer: calibration for reproducibility across echo times medRxiv. Radiol. Imaging Pub Date : 2024-01-26 Karoline Kallis, Christopher Charles Conlin, Courtney Ollison, Michael E. Hahn, Rebecca Pankow-Penner, Anders M. Dale, Tyler M. Seibert
Background: Restriction Spectrum Imaging restriction score (RSIrs) is a quantitative biomarker for detecting clinically significant prostate cancer (csPCa). However, the quantitative value of the RSIrs is affected by imaging parameters such as echo time (TE). Purpose: The purpose of the present study is to develop a calibration method to account for differences in echo times and facilitate use of RSIrs
-
A comparison of knowledge-based dose prediction approaches to assessing head and neck radiotherapy plan quality medRxiv. Radiol. Imaging Pub Date : 2024-01-26 Alexandra Olivia Leone, Mary P. Gronberg, Skylar S. Gay, Pavel A. Govyadinov, Beth Beadle, Tze Y. Lim, Thomas J. Whitaker, Karen Hoffman, Laurence E. Court, Wenhua Cao
PURPOSE: Recent studies demonstrate deep learning dose prediction algorithms may produce results like those of traditional knowledge-based planning tools. In this exploratory study, we compared 2D DVH-based knowledge-based planning tools and METHODS: Pre-validated 2D and 3D dose prediction models were applied to 58 patients with head and neck cancer treated under RTOG 0522 obtained from The Cancer
-
AI-enabled Left Atrial Volumetry in Cardiac CT Scans Improves CHARGE-AF and Outperforms NT-ProBNP for Prediction of Atrial Fibrillation in Asymptomatic Individuals: Multi-Ethnic Study of Atherosclerosis medRxiv. Radiol. Imaging Pub Date : 2024-01-24 Morteza Naghavi, David Yankelevitz, Anthony P Reeves, Matthew J Budoff, Dong Li, Kyle Atlas, Chenyu Zhang, Thomas Atlas, Daniel Levy, Jakob Wasserthal, Seth Lirette, Claudia Henschke, Christopher Defilippi, Susan Heckbert, Philip Greenland
Background: Coronary artery calcium (CAC) scans contain actionable information beyond CAC scores that is not currently reported. Methods: We have applied artificial intelligence-enabled automated cardiac chambers volumetry to CAC scans (AI-CAC) of 5535 asymptomatic individuals (52.2% women, ages 45-84) that were previously obtained for CAC scoring in the baseline examination (2000-2002) of the Multi-Ethnic
-
Dual-domain joint learning reconstruction method (JLRM) combined with physical process for spectral computed tomography medRxiv. Radiol. Imaging Pub Date : 2024-01-23 Genwei Ma, Xing Zhao
Spectral computed tomography (SCT) is an powerful imaging modality with broad applications and advantages such as contrast enhancement, artifact reduction, and material differentiation. The positive process or data collected process of SCT is a nonlinear physical process existing scatter and noise, which make it is an extremely ill-posed inverse problem in mathematics. In this paper, we propose a dual-domain
-
Comprehensive Benchmarking of CNN-Based Tumor Segmentation Methods Using Multimodal MRI Data medRxiv. Radiol. Imaging Pub Date : 2024-01-23 Kavita Kundal, K Venkateswara Rao, Arunabha Majumdar, Neeraj Kumar, Rahul Kumar
Magnetic resonance imaging (MRI) is become an essential and a frontline technique in the detection of brain tumor. However, manual segmentation of tumor from MRI scans is a time-consuming and labour-intensive process. There is a prevalent trend in employing fully automated methods for accurate tumor segmentation using MRI scans. The precision in brain tumor segmentation is essential for the better
-
Characterizing patterns of DTI variance in aging brains medRxiv. Radiol. Imaging Pub Date : 2024-01-22 Chenyu Gao, Qi Yang, Michael E. Kim, Nazirah Mohd Khairi, Leon Y. Cai, Nancy R. Newlin, Praitayini Kanakaraj, Lucas W. Remedios, Aravind R. Krishnan, Xin Yu, Tianyuan Yao, Panpan Zhang, Kurt G. Schilling, Daniel Moyer, Derek B. Archer, Susan M. Resnick, Bennett A. Landman
Background: As large analyses merge data across sites, a deeper understanding of variance in statistical assessment across the sources of data becomes critical for valid analyses. Diffusion tensor imaging (DTI) exhibits spatially varying and correlated noise, so care must be taken with distributional assumptions. Purpose: We characterize the role of physiology, subject compliance, and the interaction
-
Enhanced Classification Performance using Deep Learning Based Segmentation for Pulmonary Embolism Detection in CT Angiography medRxiv. Radiol. Imaging Pub Date : 2024-01-22 Ali Teymur Kahraman, Tomas Fröding, Dimitrios Toumpanakis, Christian Jamtheim Gustafsson, Tobias Sjöblom
Objectives: To develop a deep learning-based algorithm that automatically and accurately classifies patients as either having pulmonary emboli or not in CT pulmonary angiography (CTPA) examinations. Materials and Methods: For model development, 700 CTPA examinations from 652 patients performed at a single institution between 2014 and 2018 were used, of which 149 examinations contained 1497 PE traced
-
Pre-operative Spine Tumor Embolization: Clinical Outcomes and Effect of Embolization Completeness medRxiv. Radiol. Imaging Pub Date : 2024-01-21 Nima Omid-Fard, Jean-Paul Salameh, Matthew DF McInnes, Charles G Fisher, Manraj KS Heran
Background and Purpose To assess the association between the impact of the completeness of pre-operative spine tumour embolization and clinical outcomes including estimated blood loss (EBL), neurological status, and complications.
-
OpenMAP-T1: A Rapid Deep Learning Approach to Parcellate 280 Anatomical Regions to Cover the Whole Brain medRxiv. Radiol. Imaging Pub Date : 2024-01-20 Kei Nishimaki, Kengo Onda, Kumpei Ikuta, Yuto Uchida, Susumu Mori, Hitoshi Iyatomi, Kenichi Oishi, the Alzheimer’s Disease Neuroimaging Initiative, the Australian Imaging Biomarkers and Lifestyle flagship study of aging
This study introduces OpenMAP-T1, a deep-learning-based method for rapid and accurate whole-brain parcellation in T1-weighted brain MRI, which aims to overcome the limitations of conventional normalization-to-atlas-based approaches and multi-atlas label-fusion (MALF) techniques. Brain image parcellation is a fundamental process in neuroscientific and clinical research, enabling a detailed analysis
-
A Comparative Study: Diagnostic Performance of ChatGPT 3.5, Google Bard, Microsoft Bing, and Radiologists in Thoracic Radiology Cases medRxiv. Radiol. Imaging Pub Date : 2024-01-20 Yasin Celal Gunes, Turay Cesur
Purpose To investigate and compare the diagnostic performance of ChatGPT 3.5, Google Bard, Microsoft Bing, and two board-certified radiologists in thoracic radiology cases published by The Society of Thoracic Radiology.
-
Radiomics Analysis for Predicting Growth of Subsolid Lung Nodules on CT medRxiv. Radiol. Imaging Pub Date : 2024-01-18 Shiny Weng, Masha Bondarenko, Gunvant Chaudhari, Arun Innaje, Terrence Chen, Brandon K.K. Fields, Jae Ho Sohn
Background Accurate identification of growing subsolid nodules is crucial for effective risk stratification and the early detection of invasive lung cancer, allowing for timely treatment while avoiding unnecessary surgery on low-risk nodules that would otherwise remain stable. The traditional method of risk stratification, which relies on qualitative visual analysis of CT scans, remains challenging
-
Cross-Center Validation of Deep Learning Model for Musculoskeletal Fracture Detection in Radiographic Imaging: A Feasibility Study medRxiv. Radiol. Imaging Pub Date : 2024-01-17 Robert Hrubý, Daniel Kvak, Jakub Dandár, Anora Atakhanova, Matěj Misař, Daniel Dufek
Fractures, often resulting from trauma, overuse, or osteoporosis, pose diagnostic challenges due to their variable clinical manifestations. To address this, we propose a deep learning-based decision support system to enhance the efficacy of fracture detection in radiographic imaging. For the purpose of our study, we utilized 720 annotated musculoskeletal (MSK) X-rays from the MURA dataset, augmented