Volume 11 Issue 4 | Journal of Medical Imaging

Journal of Medical Imaging

VOL. 11 · NO. 4 | July 2024

CONTENTS

IN THIS ISSUE

Editorial (1)

Physics of Medical Imaging (4)

Image Processing (8)

Computer-Aided Diagnosis (8)

Image-Guided Procedures, Robotic Interventions, and Modeling (1)

Image Perception, Observer Performance, and Technology Assessment (4)

Biomedical Applications in Molecular, Structural, and Functional Imaging (1)

Digital Pathology (1)

< Previous Issue | Next Issue >

Receive Email Alerts

VIEW ALL ABSTRACTS +

Editorial

Highlights from JMI Issue 4

Bennett Landman

Journal of Medical Imaging, Vol. 11, Issue 04, 040101, (August 2024) https://doi.org/10.1117/1.JMI.11.4.040101

TOPICS: Artificial intelligence, Medical imaging, Computed tomography, Signal detection, Radioisotopes, Magnetic resonance imaging, Lung, Computer aided detection, Biomedical optics, Biomedical applications

Read Abstract +

Physics of Medical Imaging

Exploring single-shot propagation and speckle based phase recovery techniques for object thickness estimation by using a polychromatic X-ray laboratory source

Diego Rosich, Margarita Chevalier, Adrián Belarra, Tatiana Alieva

Journal of Medical Imaging, Vol. 11, Issue 04, 043501, (July 2024) https://doi.org/10.1117/1.JMI.11.4.043501

TOPICS: X-rays, Phase recovery, Diffusers, Speckle, X-ray sources, Breast, Adipose tissue, Biological samples, Printing, Absorption

Read Abstract +

Assessment of image quality and establishment of local acceptable quality dose for computed tomography based on patient effective diameter

Nada Hasan, Chadia Rizk, Fatema Marzooq, Khalid Khan, Maryam AlKhaja, Esameldeen Babikir

Journal of Medical Imaging, Vol. 11, Issue 04, 043502, (August 2024) https://doi.org/10.1117/1.JMI.11.4.043502

TOPICS: Computed tomography, Scanners, Image quality, Head, Body scanners, Digital Light Processing, Diagnostics, Statistical analysis, Reconstruction algorithms, Picture Archiving and Communication System

Read Abstract +

Cone-beam CT with a noncircular (sine-on-sphere) orbit: imaging performance of a clinical system for image-guided interventions

A. Kyle Jones, Moiz Ahmad, Shaan Raza, Stephen Chen, Jeffrey H. Siewerdsen

Journal of Medical Imaging, Vol. 11, Issue 04, 043503, (August 2024) https://doi.org/10.1117/1.JMI.11.4.043503

TOPICS: Modulation transfer functions, Imaging systems, Cone beam computed tomography, Computed tomography, Image quality, Head, Spatial resolution, Skull, Reconstruction algorithms, Voxels

Read Abstract +

Purpose

We aim to compare the imaging performance of a cone-beam CT (CBCT) imaging system with noncircular scan protocols (sine-on-sphere) to a conventional circular orbit.

Approach

A biplane C-arm system (ARTIS Icono; Siemens Healthineers) capable of circular and noncircular CBCT acquisition was used, with the latter orbit (sine-on-sphere, “Sine Spin”) executing a sinusoidal motion with ±10 deg tilt amplitude over the half-scan orbit. A test phantom was used for the characterization of image uniformity, noise, noise-power spectrum (NPS), spatial resolution [modulation transfer function (MTF) in axial and oblique directions], and cone-beam artifacts. Findings were interpreted using an anthropomorphic head phantom with respect to pertinent tasks in skull base neurosurgery.

Results

The noncircular scan protocol exhibited several advantages associated with improved 3D sampling—evident in the NPS as filling of the null cone about the fz spatial frequency axis and reduction of cone-beam artifacts. The region of support at the longitudinal extrema was reduced from 16 to ∼12 cm at a radial distance of 6.5 cm. Circular and noncircular orbits exhibited nearly identical image uniformity and quantum noise, demonstrating cupping of −16.7% and overall noise of ∼27 HU. Although both the radially averaged axial MTF (fx,y) and 45 deg oblique MTF (fx,y,z) were ∼20% lower for the noncircular orbit compared with the circular orbit at the default full reconstruction field of view (FOV), there was no difference in spatial resolution for the medium reconstruction FOV (smaller voxel size). Differences in the perceptual image quality for the anthropomorphic phantom reinforced the objective, quantitative findings, including reduced beam-hardening and cone-beam artifacts about structures of interest in the skull base.

Conclusions

Image quality differences between circular and noncircular CBCT orbits were quantitatively evaluated on a clinical system in the context of neurosurgery. The primary performance advantage for the noncircular orbit was the improved sampling and elimination of cone-beam artifacts.

CT reconstruction using diffusion posterior sampling conditioned on a nonlinear measurement model

Shudong Li, Xiao Jiang, Matthew Tivnan, Grace J. Gang, Yuan Shen, J. Webster Stayman

Journal of Medical Imaging, Vol. 11, Issue 04, 043504, (August 2024) https://doi.org/10.1117/1.JMI.11.4.043504

TOPICS: Double positive medium, Diffusion, CT reconstruction, Education and training, Data modeling, Systems modeling, Reverse modeling, Lung, Computed tomography, Error analysis

Read Abstract +

Image Processing

Lung vessel connectivity map as anatomical prior knowledge for deep learning-based lung lobe segmentation

Simone Bendazzoli, Emelie Bäcklin, Örjan Smedby, Birgitta Janerot-Sjoberg, Bryan Connolly, Chunliang Wang

Journal of Medical Imaging, Vol. 11, Issue 04, 044001, (July 2024) https://doi.org/10.1117/1.JMI.11.4.044001

TOPICS: Lung, Image segmentation, Anatomy, Education and training, COVID 19, Prior knowledge, Voxels, Performance modeling, Computed tomography, Network architectures

Read Abstract +

Projected pooling loss for red nucleus segmentation with soft topology constraints

Guanghui Fu, Rosana El Jurdi, Lydia Chougar, Didier Dormont, Romain Valabregue, Stéphane Lehéricy, Olivier Colliot, The ICEBERG Study Group

Journal of Medical Imaging, Vol. 11, Issue 04, 044002, (July 2024) https://doi.org/10.1117/1.JMI.11.4.044002

TOPICS: Image segmentation, Education and training, Anatomy, Voxels, Optical spheres, Spleen, Prior knowledge, Deep learning, Medical imaging, Heart

Read Abstract +

Optimizing neurointerventional procedures: an algorithm for embolization coil detection and automated collimation to enable dose reduction

Arpitha Ravi, Philipp Bernhardt, Mathis Hoffmann, Richard Obler, Cuong Nguyen, Andreas Berting, René Chapot, Andreas Maier

Journal of Medical Imaging, Vol. 11, Issue 04, 044003, (July 2024) https://doi.org/10.1117/1.JMI.11.4.044003

TOPICS: Collimation, Data modeling, Object detection, Blob detection, Sensors, Education and training, Deep learning, Mathematical optimization, X-rays, Image processing

Read Abstract +

Purpose

Monitoring radiation dose and time parameters during radiological interventions is crucial, especially in neurointerventional procedures, such as aneurysm treatment with embolization coils. The algorithm presented detects the presence of these embolization coils in medical images. It establishes a bounding box as a reference for automated collimation, with the primary objective being to enhance the efficiency and safety of neurointerventional procedures by actively optimizing image quality while minimizing patient dose.

Methods

Two distinct methodologies are evaluated in our study. The first involves deep learning, employing the Faster R-CNN model with a ResNet-50 FPN as a backbone and a RetinaNet model. The second method utilizes a classical blob detection approach, serving as a benchmark for comparison.

Results

We performed a fivefold cross-validation, and our top-performing model achieved mean mAP@75 of 0.84 across all folds on validation data and mean mAP@75 of 0.94 on independent test data. Since we use an upscaled bounding box, achieving 100% overlap between ground truth and prediction is not necessary. To highlight the real-world applications of our algorithm, we conducted a simulation featuring a coil constructed from an alloy wire, effectively showcasing the implementation of automatic collimation. This resulted in a notable reduction in the dose area product, signifying the reduction of stochastic risks for both patients and medical staff by minimizing scatter radiation. Additionally, our algorithm assists in avoiding extreme brightness or darkness in X-ray angiography images during narrow collimation, ultimately streamlining the collimation process for physicians.

Conclusion

To our knowledge, this marks the initial attempt at an approach successfully detecting embolization coils, showcasing the extended applications of integrating detection results into the X-ray angiography system. The method we present has the potential for broader application, allowing its extension to detect other medical objects utilized in interventional procedures.

Left ventricular structural integrity on tetralogy of Fallot patients: approach using longitudinal relaxation time mapping

Giorgos Broumpoulis, Efstratios Karavasilis, Niki Lama, Ioannis Papadopoulos, Panagiotis Zachos, Sotiria Apostolopoulou, Nikolaos Kelekis

Journal of Medical Imaging, Vol. 11, Issue 04, 044004, (August 2024) https://doi.org/10.1117/1.JMI.11.4.044004

TOPICS: Cardiovascular magnetic resonance imaging, Magnetic resonance imaging, Heart, Gadolinium, Statistical analysis, Cardiovascular disorders, Tissues, Matrices, Arteries, Visualization

Read Abstract +

Dose robustness of deep learning models for anatomic segmentation of computed tomography images

Artyom Tsanda, Hannes Nickisch, Tobias Wissel, Tobias Klinder, Tobias Knopp, Michael Grass

Journal of Medical Imaging, Vol. 11, Issue 04, 044005, (August 2024) https://doi.org/10.1117/1.JMI.11.4.044005

TOPICS: Image segmentation, Denoising, Computed tomography, Data modeling, Education and training, Deep learning, Anatomy, Simulations, 3D modeling, Semantics

Read Abstract +

Stain SAN: simultaneous augmentation and normalization for histopathology images

Taebin Kim, Yao Li, Benjamin Calhoun, Aatish Thennavan, Lisa A. Carey, W. Fraser Symmans, Melissa Troester, Charles Perou, J.S. Marron

Journal of Medical Imaging, Vol. 11, Issue 04, 044006, (August 2024) https://doi.org/10.1117/1.JMI.11.4.044006

TOPICS: Education and training, Matrices, Histopathology, Color, Binary data, Color normalization, RGB color model, Image classification, Batch normalization, Visualization

Read Abstract +

Purpose

We address the need for effective stain domain adaptation methods in histopathology to enhance the performance of downstream computational tasks, particularly classification. Existing methods exhibit varying strengths and weaknesses, prompting the exploration of a different approach. The focus is on improving stain color consistency, expanding the stain domain scope, and minimizing the domain gap between image batches.

Approach

We introduce a new domain adaptation method, Stain simultaneous augmentation and normalization (SAN), designed to adjust the distribution of stain colors to align with a target distribution. Stain SAN combines the merits of established methods, such as stain normalization, stain augmentation, and stain mix-up, while mitigating their inherent limitations. Stain SAN adapts stain domains by resampling stain color matrices from a well-structured target distribution.

Results

Experimental evaluations of cross-dataset clinical estrogen receptor status classification demonstrate the efficacy of Stain SAN and its superior performance compared with existing stain adaptation methods. In one case, the area under the curve (AUC) increased by 11.4%. Overall, our results clearly show the improvements made over the history of the development of these methods culminating with substantial enhancement provided by Stain SAN. Furthermore, we show that Stain SAN achieves results comparable with the state-of-the-art generative adversarial network-based approach without requiring separate training for stain adaptation or access to the target domain during training. Stain SAN’s performance is on par with HistAuGAN, proving its effectiveness and computational efficiency.

Conclusions

Stain SAN emerges as a promising solution, addressing the potential shortcomings of contemporary stain adaptation methods. Its effectiveness is underscored by notable improvements in the context of clinical estrogen receptor status classification, where it achieves the best AUC performance. The findings endorse Stain SAN as a robust approach for stain domain adaptation in histopathology images, with implications for advancing computational tasks in the field.

Characterizing patterns of diffusion tensor imaging variance in aging brains

Chenyu Gao, Qi Yang, Michael Kim, Nazirah Mohd Khairi, Leon Cai, Nancy Newlin, Praitayini Kanakaraj, Lucas Remedios, Aravind Krishnan, Xin Yu, Tianyuan Yao, Panpan Zhang, Kurt Schilling, Daniel Moyer, Derek Archer, Susan Resnick, Bennett Landman, for the Alzheimer’s Disease Neuroimaging Initiative, The BIOCARD Study team

Journal of Medical Imaging, Vol. 11, Issue 04, 044007, (August 2024) https://doi.org/10.1117/1.JMI.11.4.044007

TOPICS: Diffusion tensor imaging, Image segmentation, Data modeling, Scanners, Brain, Neuroimaging, Motion models, Signal to noise ratio, Matrices, Head

Read Abstract +

Purpose

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. Here, we characterize the role of physiology, subject compliance, and the interaction of the subject with the scanner in the understanding of DTI variability, as modeled in the spatial variance of derived metrics in homogeneous regions.

Approach

We analyze DTI data from 1035 subjects in the Baltimore Longitudinal Study of Aging, with ages ranging from 22.4 to 103 years old. For each subject, up to 12 longitudinal sessions were conducted. We assess the variance of DTI scalars within regions of interest (ROIs) defined by four segmentation methods and investigate the relationships between the variance and covariates, including baseline age, time from the baseline (referred to as “interval”), motion, sex, and whether it is the first scan or the second scan in the session.

Results

Covariate effects are heterogeneous and bilaterally symmetric across ROIs. Inter-session interval is positively related (p≪0.001) to FA variance in the cuneus and occipital gyrus, but negatively (p≪0.001) in the caudate nucleus. Males show significantly (p≪0.001) higher FA variance in the right putamen, thalamus, body of the corpus callosum, and cingulate gyrus. In 62 out of 176 ROIs defined by the Eve type-1 atlas, an increase in motion is associated (p<0.05) with a decrease in FA variance. Head motion increases during the rescan of DTI (Δμ=0.045 mm per volume).

Conclusions

The effects of each covariate on DTI variance and their relationships across ROIs are complex. Ultimately, we encourage researchers to include estimates of variance when sharing data and consider models of heteroscedasticity in analysis. This work provides a foundation for study planning to account for regional variations in metric variance.

Field-of-view extension for brain diffusion MRI via deep generative models

Chenyu Gao, Shunxing Bao, Michael Kim, Nancy Newlin, Praitayini Kanakaraj, Tianyuan Yao, Gaurav Rudravaram, Yuankai Huo, Daniel Moyer, Kurt Schilling, Walter Kukull, Arthur Toga, Derek Archer, Timothy Hohman, Bennett Landman, Zhiyuan Li

Journal of Medical Imaging, Vol. 11, Issue 04, 044008, (August 2024) https://doi.org/10.1117/1.JMI.11.4.044008

TOPICS: Brain, Diffusion weighted imaging, Neuroimaging, Diffusion magnetic resonance imaging, Education and training, Alzheimer disease, Diffusion, Data modeling, Anatomy, Medical imaging

Read Abstract +

Purpose

In brain diffusion magnetic resonance imaging (dMRI), the volumetric and bundle analyses of whole-brain tissue microstructure and connectivity can be severely impeded by an incomplete field of view (FOV). We aim to develop a method for imputing the missing slices directly from existing dMRI scans with an incomplete FOV. We hypothesize that the imputed image with a complete FOV can improve whole-brain tractography for corrupted data with an incomplete FOV. Therefore, our approach provides a desirable alternative to discarding the valuable brain dMRI data, enabling subsequent tractography analyses that would otherwise be challenging or unattainable with corrupted data.

Approach

We propose a framework based on a deep generative model that estimates the absent brain regions in dMRI scans with an incomplete FOV. The model is capable of learning both the diffusion characteristics in diffusion-weighted images (DWIs) and the anatomical features evident in the corresponding structural images for efficiently imputing missing slices of DWIs in the incomplete part of the FOV.

Results

For evaluating the imputed slices, on the Wisconsin Registry for Alzheimer’s Prevention (WRAP) dataset, the proposed framework achieved PSNRb0=22.397, SSIMb0=0.905, PSNRb1300=22.479, and SSIMb1300=0.893; on the National Alzheimer’s Coordinating Center (NACC) dataset, it achieved PSNRb0=21.304, SSIMb0=0.892, PSNRb1300=21.599, and SSIMb1300=0.877. The proposed framework improved the tractography accuracy, as demonstrated by an increased average Dice score for 72 tracts (p<0.001) on both the WRAP and NACC datasets.

Conclusions

Results suggest that the proposed framework achieved sufficient imputation performance in brain dMRI data with an incomplete FOV for improving whole-brain tractography, thereby repairing the corrupted data. Our approach achieved more accurate whole-brain tractography results with an extended and complete FOV and reduced the uncertainty when analyzing bundles associated with Alzheimer’s disease.

Computer-Aided Diagnosis

Examining feature extraction and classification modules in machine learning for diagnosis of low-dose computed tomographic screening-detected in vivo lesions

Daniel Liang, David Liang, Marc J. Pomeroy, Yongfeng Gao, Licheng Kuo, Lihong Li

Journal of Medical Imaging, Vol. 11, Issue 04, 044501, (July 2024) https://doi.org/10.1117/1.JMI.11.4.044501

TOPICS: Feature extraction, Image classification, Computed tomography, Tissues, Polyps, Machine learning, In vivo imaging, X-ray computed tomography, Medical imaging, Computer aided detection

Read Abstract +

Purpose

Medical imaging-based machine learning (ML) for computer-aided diagnosis of in vivo lesions consists of two basic components or modules of (i) feature extraction from non-invasively acquired medical images and (ii) feature classification for prediction of malignancy of lesions detected or localized in the medical images. This study investigates their individual performances for diagnosis of low-dose computed tomography (CT) screening-detected lesions of pulmonary nodules and colorectal polyps.

Approach

Three feature extraction methods were investigated. One uses the mathematical descriptor of gray-level co-occurrence image texture measure to extract the Haralick image texture features (HFs). One uses the convolutional neural network (CNN) architecture to extract deep learning (DL) image abstractive features (DFs). The third one uses the interactions between lesion tissues and X-ray energy of CT to extract tissue-energy specific characteristic features (TFs). All the above three categories of extracted features were classified by the random forest (RF) classifier with comparison to the DL-CNN method, which reads the images, extracts the DFs, and classifies the DFs in an end-to-end manner. The ML diagnosis of lesions or prediction of lesion malignancy was measured by the area under the receiver operating characteristic curve (AUC). Three lesion image datasets were used. The lesions’ tissue pathological reports were used as the learning labels.

Results

Experiments on the three datasets produced AUC values of 0.724 to 0.878 for the HFs, 0.652 to 0.965 for the DFs, and 0.985 to 0.996 for the TFs, compared to the DL-CNN of 0.694 to 0.964. These experimental outcomes indicate that the RF classifier performed comparably to the DL-CNN classification module and the extraction of tissue-energy specific characteristic features dramatically improved AUC value.

Conclusions

The feature extraction module is more important than the feature classification module. Extraction of tissue-energy specific characteristic features is more important than extraction of image abstractive and characteristic features.

Pulmonary nodule detection in low dose computed tomography using a medical-to-medical transfer learning approach

Jenita Manokaran, Richa Mittal, Eranga Ukwatta

Journal of Medical Imaging, Vol. 11, Issue 04, 044502, (July 2024) https://doi.org/10.1117/1.JMI.11.4.044502

TOPICS: Cancer detection, Object detection, Education and training, Lung cancer, Tumor growth modeling, Lung, Data modeling, Computed tomography, Performance modeling, Deep learning

Read Abstract +

Purpose

Lung cancer is the second most common cancer and the leading cause of cancer death globally. Low dose computed tomography (LDCT) is the recommended imaging screening tool for the early detection of lung cancer. A fully automated computer-aided detection method for LDCT will greatly improve the existing clinical workflow. Most of the existing methods for lung detection are designed for high-dose CTs (HDCTs), and those methods cannot be directly applied to LDCTs due to domain shifts and inferior quality of LDCT images. In this work, we describe a semi-automated transfer learning-based approach for the early detection of lung nodules using LDCTs.

Approach

In this work, we developed an algorithm based on the object detection model, you only look once (YOLO) to detect lung nodules. The YOLO model was first trained on CTs, and the pre-trained weights were used as initial weights during the retraining of the model on LDCTs using a medical-to-medical transfer learning approach. The dataset for this study was from a screening trial consisting of LDCTs acquired from 50 biopsy-confirmed lung cancer patients obtained over 3 consecutive years (T1, T2, and T3). About 60 lung cancer patients’ HDCTs were obtained from a public dataset. The developed model was evaluated using a hold-out test set comprising 15 patient cases (93 slices with cancerous nodules) using precision, specificity, recall, and F1-score. The evaluation metrics were reported patient-wise on a per-year basis and averaged for 3 years. For comparative analysis, the proposed detection model was trained using pre-trained weights from the COCO dataset as the initial weights. A paired t-test and chi-squared test with an alpha value of 0.05 were used for statistical significance testing.

Results

The results were reported by comparing the proposed model developed using HDCT pre-trained weights with COCO pre-trained weights. The former approach versus the latter approach obtained a precision of 0.982 versus 0.93 in detecting cancerous nodules, specificity of 0.923 versus 0.849 in identifying slices with no cancerous nodules, recall of 0.87 versus 0.886, and F1-score of 0.924 versus 0.903. As the nodule progressed, the former approach achieved a precision of 1, specificity of 0.92, and sensitivity of 0.930. The statistical analysis performed in the comparative study resulted in a p-value of 0.0054 for precision and a p-value of 0.00034 for specificity.

Conclusions

In this study, a semi-automated method was developed to detect lung nodules in LDCTs using HDCT pre-trained weights as the initial weights and retraining the model. Further, the results were compared by replacing HDCT pre-trained weights in the above approach with COCO pre-trained weights. The proposed method may identify early lung nodules during the screening program, reduce overdiagnosis and follow-ups due to misdiagnosis in LDCTs, start treatment options in the affected patients, and lower the mortality rate.

Learning carotid vessel wall segmentation in black-blood MRI using sparsely sampled cross-sections from 3D data

Hinrich Rahlfs, Markus Hüllebrand, Sebastian Schmitter, Christoph Strecker, Andreas Harloff, Anja Hennemuth

Journal of Medical Imaging, Vol. 11, Issue 04, 044503, (July 2024) https://doi.org/10.1117/1.JMI.11.4.044503

TOPICS: Image segmentation, Arteries, Education and training, Magnetic resonance imaging, Data modeling, Independent component analysis, Simulation of CCA and DLA aggregates, 3D image processing, Scanners, Performance modeling

Read Abstract +

Learning three-dimensional aortic root assessment based on sparse annotations

Johanna Brosig, Nina Krüger, Inna Khasyanova, Isaac Wamala, Matthias Ivantsits, Simon Sündermann, Jörg Kempfert, Stefan Heldmann, Anja Hennemuth

Journal of Medical Imaging, Vol. 11, Issue 04, 044504, (July 2024) https://doi.org/10.1117/1.JMI.11.4.044504

TOPICS: 3D mask effects, Image segmentation, Education and training, 3D modeling, Computed tomography, Anatomy, Risk assessment, Reproducibility, Aorta, Neural networks

Read Abstract +

AI-based automated segmentation for ovarian/adnexal masses and their internal components on ultrasound imaging

Heather Whitney, Roni Yoeli-Bik, Jacques Abramowicz, Li Lan, Hui Li, Ryan Longman, Ernst Lengyel, Maryellen Giger

Journal of Medical Imaging, Vol. 11, Issue 04, 044505, (August 2024) https://doi.org/10.1117/1.JMI.11.4.044505

TOPICS: Image segmentation, Ultrasonography, Tissues, Education and training, Ovarian cancer, Radiomics, Pathology, Tumor growth modeling, Feature extraction, Solids

Read Abstract +

Capability and reliability of deep learning models to make density predictions on low-dose mammograms

Steven Squires, Alistair Mackenzie, Dafydd Gareth Evans, Sacha Howell, Susan Astley

Journal of Medical Imaging, Vol. 11, Issue 04, 044506, (August 2024) https://doi.org/10.1117/1.JMI.11.4.044506

TOPICS: Breast, Breast density, Mammography, Data modeling, Education and training, Performance modeling, Deep learning, Reliability, Histograms, Cancer

Read Abstract +

Exploring synthetic datasets for computer-aided detection: a case study using phantom scan data for enhanced lung nodule false positive reduction

Mohammad Mehdi Farhangi, Michael Maynord, Cornelia Fermüller, Yiannis Aloimonos, Berkman Sahiner, Nicholas Petrick

Journal of Medical Imaging, Vol. 11, Issue 04, 044507, (August 2024) https://doi.org/10.1117/1.JMI.11.4.044507

TOPICS: Education and training, Computed tomography, Computer aided detection, Databases, Data modeling, Neural networks, Machine learning, Lung, Tunable filters, Detection and tracking algorithms

Read Abstract +

Purpose

Synthetic datasets hold the potential to offer cost-effective alternatives to clinical data, ensuring privacy protections and potentially addressing biases in clinical data. We present a method leveraging such datasets to train a machine learning algorithm applied as part of a computer-aided detection (CADe) system.

Approach

Our proposed approach utilizes clinically acquired computed tomography (CT) scans of a physical anthropomorphic phantom into which manufactured lesions were inserted to train a machine learning algorithm. We treated the training database obtained from the anthropomorphic phantom as a simplified representation of clinical data and increased the variability in this dataset using a set of randomized and parameterized augmentations. Furthermore, to mitigate the inherent differences between phantom and clinical datasets, we investigated adding unlabeled clinical data into the training pipeline.

Results

We apply our proposed method to the false positive reduction stage of a lung nodule CADe system in CT scans, in which regions of interest containing potential lesions are classified as nodule or non-nodule regions. Experimental results demonstrate the effectiveness of the proposed method; the system trained on labeled data from physical phantom scans and unlabeled clinical data achieves a sensitivity of 90% at eight false positives per scan. Furthermore, the experimental results demonstrate the benefit of the physical phantom in which the performance in terms of competitive performance metric increased by 6% when a training set consisting of 50 clinical CT scans was enlarged by the scans obtained from the physical phantom.

Conclusions

The scalability of synthetic datasets can lead to improved CADe performance, particularly in scenarios in which the size of the labeled clinical data is limited or subject to inherent bias. Our proposed approach demonstrates an effective utilization of synthetic datasets for training machine learning algorithms.

Generative adversarial network–based reconstruction of healthy anatomy for anomaly detection in brain CT scans

Sina Walluscheck, Annika Gerken, Ivana Galinovic, Kersten Villringer, Jochen Fiebach, Jan Klein, Stefan Heldmann

Journal of Medical Imaging, Vol. 11, Issue 04, 044508, (August 2024) https://doi.org/10.1117/1.JMI.11.4.044508

TOPICS: Education and training, Computed tomography, Brain, Image restoration, Neuroimaging, Medical image reconstruction, Gallium nitride, Anatomy, Image segmentation, Tumors

Read Abstract +

Image-Guided Procedures, Robotic Interventions, and Modeling

Improving radiological quantification of levator hiatus features with measures informed by statistical shape modeling

Vincenzia Vargo, Megan Routzong, Pamela Moalli, Ghazaleh Rostaminia, Steven Abramowitch

Journal of Medical Imaging, Vol. 11, Issue 04, 045001, (August 2024) https://doi.org/10.1117/1.JMI.11.4.045001

TOPICS: Muscles, Modeling, Image segmentation, Magnetic resonance imaging, Anatomy, Shape analysis, Visualization, Diseases and disorders, Deformation, Cross validation

Read Abstract +

Image Perception, Observer Performance, and Technology Assessment

Greater benefits of deep learning-based computer-aided detection systems for finding small signals in 3D volumetric medical images

Devi Klein, Srijita Karmakar, Aditya Jonnalagadda, Craig Abbey, Miguel Eckstein

Journal of Medical Imaging, Vol. 11, Issue 04, 045501, (July 2024) https://doi.org/10.1117/1.JMI.11.4.045501

TOPICS: Signal detection, Computer aided detection, Digital breast tomosynthesis, 3D modeling, 3D image processing, Eye, Visualization, Education and training, Deep learning, Picture Archiving and Communication System

Read Abstract +

Comparison of human observer impression of X-ray fluoroscopy and angiography image quality with technical changes to image quality

Jelena Mihailovic, Yoshihisa Kanaji, Daniel Miller, Malcolm Bell, Kenneth Fetterly

Journal of Medical Imaging, Vol. 11, Issue 04, 045502, (August 2024) https://doi.org/10.1117/1.JMI.11.4.045502

TOPICS: Image quality, Sensors, Fluoroscopy, X-rays, Video, Angiography, Visualization, Image sharpness, Quantum noise, Arteries

Read Abstract +

Purpose

Spatio-temporal variability in clinical fluoroscopy and cine angiography images combined with nonlinear image processing prevents the application of traditional image quality measurements in the cardiac catheterization laboratory. We aimed to develop and validate methods to measure human observer impressions of the image quality.

Approach

Multi-frame images of the thorax of a euthanized pig were acquired to provide an anatomical background. The detector dose was varied from 6 to 200 nGy (increments 2×), and 0.6 and 1.0 mm focal spots were used. Two coronary stents with/without 0.5 mm separation and a synthetic right coronary artery (RCA) with hemispherical defects were embedded into the background images as test objects. The quantitative observer (n=17) performance was measured using a two-alternating forced-choice test of whether stents were separated and by a count of visible right coronary artery defects. Qualitative impressions of noise, spatial resolution, and overall image quality were measured using a visual analog scale (VAS). A paired t-test and multinomial logistic regression model were used to identify statistically significant factors affecting the observer’s impression image quality.

Results

The proportion of correct detection of stent separation and the number of reported right coronary artery defects changed significantly with detector dose increment in the 6 to 100 nGy (p<0.05). Although a trend favored the 0.6 versus 1.0 mm focal spot for these quantitative assessments, this was insignificant. Visual analog scale measurements changed significantly with detector dose increments in the range of 24 to 100 nGy and focal spot size (p<0.05). The application of multinomial logistic regression analysis to observer VAS scores demonstrated sensitivity matching of the paired t-test applied to quantitative observer performance measurements.

Conclusions

Both quantitative and qualitative measurements of observer impression of the image quality were sensitive to image quality changes associated with changing the detector dose and focal spot size. These findings encourage future work that uses qualitative image quality measurements to assess clinical fluoroscopy and angiography image quality.

Task-based assessment for neural networks: evaluating undersampled MRI reconstructions based on human observer signal detection

Joshua Herman, Rachel Roca, Alexandra O'Neill, Marcus Wong, Sajan Goud Lingala, Angel Pineda

Journal of Medical Imaging, Vol. 11, Issue 04, 045503, (August 2024) https://doi.org/10.1117/1.JMI.11.4.045503

TOPICS: Education and training, Neural networks, Image quality, Signal detection, Image restoration, Signal attenuation, Magnetic resonance imaging, Cross validation, Network architectures, Statistical analysis

Read Abstract +

Use of reporting templates for chest radiographs in a coronavirus disease 2019 context: measuring concordance of radiologists with three international templates

Sarah Lewis, Jayden Wells, Warren M Reed, Claudia Mello-Thoms, Peter O’Reilly, Marion Dimigen

Journal of Medical Imaging, Vol. 11, Issue 04, 045504, (August 2024) https://doi.org/10.1117/1.JMI.11.4.045504

TOPICS: COVID 19, Chest imaging, Radiology, Image quality, Diagnostics, Radiography, Pulmonary disorders, Medicine, Computed tomography, Pathology

Read Abstract +

Purpose

Reporting templates for chest radiographs (CXRs) for patients presenting or being clinically managed for severe acute respiratory syndrome coronavirus 2 [coronavirus disease 2019 (COVID-19)] has attracted advocacy from international radiology societies. We aim to explore the effectiveness and useability of three international templates through the concordance of, and between, radiologists reporting on the presence and severity of COVID-19 on CXRs.

Approach

Seventy CXRs were obtained from a referral hospital, 50 from patients with COVID-19 (30 rated “classic” COVID-19 appearance and 20 “indeterminate”) and 10 “normal” and 10 “alternative pathology” CXRs. The recruited radiologists were assigned to three test sets with the same CXRs but with different template orders. Each radiologist read their test set three times and assigned a classification to the CXR using the Royal Australian New Zealand College of Radiology (RANZCR), British Society of Thoracic Imaging (BSTI), and Modified COVID-19 Reporting and Data System (Dutch; mCO-RADS) templates. Inter-reader variability and intra-reader variability were measured using Fleiss’ kappa coefficient.

Results

Twelve Australian radiologists participated. The BSTI template had the highest inter-reader agreement (0.46; “moderate” agreement), followed by RANZCR (0.45) and mCO-RADS (0.32). Concordance was driven by strong agreement in “normal” and “alternative” classifications and was lowest for “indeterminate.” General consistency was observed across classifications and templates, with intra-reader variability ranging from “good” to “very good” for COVID-19 CXRs (0.61), “normal” CXRs (0.76), and “alternative” (0.68).

Conclusions

Reporting templates may be useful in reducing variation among radiology reports, with intra-reader variability showing promise. Feasibility and implementation require a wider approach including referring and treating doctors plus the development of training packages for radiologists specific to the template being used.

Biomedical Applications in Molecular, Structural, and Functional Imaging

Machine learning and magnetic resonance image texture analysis predicts accelerated lung function decline in ex-smokers with and without chronic obstructive pulmonary disease

Maksym Sharma, Miranda Kirby, Aaron Fenster, David McCormack, Grace Parraga

Journal of Medical Imaging, Vol. 11, Issue 04, 046001, (July 2024) https://doi.org/10.1117/1.JMI.11.4.046001

TOPICS: Magnetic resonance imaging, Lung, Machine learning, Image analysis, Education and training, Optical spheres, Chronic obstructive pulmonary disease, Magnetism, Cooccurrence matrices, Voxels

Read Abstract +

Purpose

Our objective was to train machine-learning algorithms on hyperpolarized He3 magnetic resonance imaging (MRI) datasets to generate models of accelerated lung function decline in participants with and without chronic-obstructive-pulmonary-disease. We hypothesized that hyperpolarized gas MRI ventilation, machine-learning, and multivariate modeling could be combined to predict clinically-relevant changes in forced expiratory volume in 1 s (FEV1) across 3 years.

Approach

Hyperpolarized He3 MRI was acquired using a coronal Cartesian fast gradient recalled echo sequence with a partial echo and segmented using a k-means clustering algorithm. A maximum entropy mask was used to generate a region-of-interest for texture feature extraction using a custom-developed algorithm and the PyRadiomics platform. The principal component and Boruta analyses were used for feature selection. Ensemble-based and single machine-learning classifiers were evaluated using area-under-the-receiver-operator-curve and sensitivity-specificity analysis.

Results

We evaluated 88 ex-smoker participants with 31±7 months follow-up data, 57 of whom (22 females/35 males, 70±9 years) had negligible changes in FEV1 and 31 participants (7 females/24 males, 68±9 years) with worsening FEV1≥60 mL/year. In addition, 3/88 ex-smokers reported a change in smoking status. We generated machine-learning models to predict FEV1 decline using demographics, spirometry, and texture features, with the later yielding the highest classification accuracy of 81%. The combined model (trained on all available measurements) achieved the overall best classification accuracy of 82%; however, it was not significantly different from the model trained on MRI texture features alone.

Conclusion

For the first time, we have employed hyperpolarized He3 MRI ventilation texture features and machine-learning to identify ex-smokers with accelerated decline in FEV1 with 82% accuracy.

Digital Pathology

Transformation from hematoxylin-and-eosin staining to Ki-67 immunohistochemistry digital staining images using deep learning: experimental validation on the labeling index

Cunyuan Ji, Kengo Oshima, Takumi Urata, Fumikazu Kimura, Keiko Ishii, Takeshi Uehara, Kenji Suzuki, Saori Takeyama, Masahiro Yamaguchi

Journal of Medical Imaging, Vol. 11, Issue 04, 047501, (July 2024) https://doi.org/10.1117/1.JMI.11.4.047501

TOPICS: RGB color model, Education and training, Lithium, Visualization, Deep learning, Cross validation, Current controlled voltage source, Tumors, Tumor growth modeling, Matrices

Read Abstract +

Purpose

Endometrial cancer (EC) is one of the most common types of cancer affecting women. While the hematoxylin-and-eosin (H&E) staining remains the standard for histological analysis, the immunohistochemistry (IHC) method provides molecular-level visualizations. Our study proposes a digital staining method to generate the hematoxylin-3,3′-diaminobenzidine (H-DAB) IHC stain of Ki-67 for the whole slide image of the EC tumor from its H&E stain counterpart.

Approach

We employed a color unmixing technique to yield stain density maps from the optical density (OD) of the stains and utilized the U-Net for end-to-end inference. The effectiveness of the proposed method was evaluated using the Pearson correlation between the digital and physical stain’s labeling index (LI), a key metric indicating tumor proliferation. Two different cross-validation schemes were designed in our study: intraslide validation and cross-case validation (CCV). In the widely used intraslide scheme, the training and validation sets might include different regions from the same slide. The rigorous CCV validation scheme strictly prohibited any validation slide from contributing to training.

Results

The proposed method yielded a high-resolution digital stain with preserved histological features, indicating a reliable correlation with the physical stain in terms of the Ki-67 LI. In the intraslide scheme, using intraslide patches resulted in a biased accuracy (e.g., R=0.98) significantly higher than that of CCV. The CCV scheme retained a fair correlation (e.g., R=0.66) between the LIs calculated from the digital stain and its physical IHC counterpart. Inferring the OD of the IHC stain from that of the H&E stain enhanced the correlation metric, outperforming that of the baseline model using the RGB space.

Conclusions

Our study revealed that molecule-level insights could be obtained from H&E images using deep learning. Furthermore, the improvement brought via OD inference indicated a possible method for creating more generalizable models for digital staining via per-stain analysis.