Medical Imaging, Quantitaive Image Analysis, Computer Aided Diagnosis, Machine Learning
Our study on cardiac toxicity in lung cancer treatment is now featured in a JCO CCI editorial. Discoveries that could change patient care are on the horizon. Stay tuned! #CardiacToxicity#LungCancer#Innovation
Our paper “Novel Functional Radiomics for Prediction of Cardiac Positron Emission Tomography Avidity in Lung Cancer Radiotherapy” has been published in JCO CCI. This research work delves into an innovative approach to predict clinical cardiac assessment using functional imaging.
Traditional methods for evaluating cardiotoxicity primarily focus on radiation doses to the heart. However, functional imaging offers the potential to enhance early prediction of cardiotoxicity in lung cancer patients undergoing radiotherapy. In this context, Fluorine-18 (18F) fluorodeoxyglucose (FDG)-positron emission tomography (PET)/computed tomography (CT) imaging plays a crucial role. This study aims to develop a radiomics model that predicts clinical cardiac assessment using 18F-FDG PET/CT scans before thoracic radiation therapy.
The results showed that out of 202 scans, cardiac FDG uptake was scored as no uptake (39.6%), diffuse uptake (25.3%), and focal uptake (35.1%). The researchers reduced sixty-two independent radiomics features to nine clinically pertinent features. The best model showed a predictive accuracy of 93% in the training data set and 80% and 92% in two external validation data sets.
This study represents a significant advancement by developing and evaluating functional cardiac radiomic features from standard-of-care FDG PET/CT scans. The results demonstrate good predictive accuracy when compared to clinical imaging evaluation.
Feel free to explore the full paper in the JCO Clinical Cancer Informatics, Volume 8, available at this link.
Related Presentations
Functional Delta-Radiomics Overall Survival Prediction
Functional Radiomics Classification of Cardiac Uptake Patterns
https://www.abstractsonline.com/pp8/#!/10856/presentation/7201
This paper has been published in the Computational and Structural Biotechnology Journal.
Wookjin Choi a d1, Chia-Ju Liu b 1, Sadegh Riyahi Alam a, Jung Hun Oh a, Raj Vaghjiani c, John Humm a, Wolfgang Weber b, Prasad S. Adusumilli c, Joseph O. Deasy a, Wei Lu a
a Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, NY 10065, USA
b Department of Radiology, Memorial Sloan Kettering Cancer Center, New York, NY 10065, USA
c Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY 10065, USA
d Department of Radiation Oncology, Thomas Jefferson University, Philadelphia, PA 19107, USA
Lung adenocarcinoma (ADC) is the most common non-small cell lung cancer. Surgical resection is the primary treatment for early-stage lung ADC while lung-sparing surgery is an alternative for non-aggressive cases. Identifying histopathologic subtypes before surgery helps determine the optimal surgical approach. Predominantly solid or micropapillary (MIP) subtypes are aggressive and associated with a higher likelihood of recurrence and metastasis and lower survival rates. This study aims to non-invasively identify these aggressive subtypes using preoperative 18F-FDG PET/CT and diagnostic CT radiomics analysis. We retrospectively studied 119 patients with stage I lung ADC and tumors ≤ 2 cm, where 23 had aggressive subtypes (18 solid and 5 MIPs). Out of 214 radiomic features from the PET/CT and CT scans and 14 clinical parameters, 78 significant features (3 CT and 75 PET features) were identified through univariate analysis and hierarchical clustering with minimized feature collinearity. A combination of Support Vector Machine classifier and Least Absolute Shrinkage and Selection Operator built predictive models. Ten iterations of 10-fold cross-validation (10 ×10-fold CV) evaluated the model. A pair of texture feature (PET GLCM Correlation) and shape feature (CT Sphericity) emerged as the best predictor. The radiomics model significantly outperformed the conventional predictor SUVmax (accuracy: 83.5% vs. 74.7%, p = 9e-9) and identified aggressive subtypes by evaluating FDG uptake in the tumor and tumor shape. It also demonstrated a high negative predictive value of 95.6% compared to SUVmax (88.2%, p = 2e-10). The proposed radiomics approach could reduce unnecessary extensive surgeries for non-aggressive subtype patients, improving surgical decision-making for early-stage lung ADC patients.
https://www.sciencedirect.com/science/article/pii/S2001037023004233
2017 AAPM annual meeting
http://www.aapm.org/meetings/2017AM/PRAbs.asp?mid=127&aid=37917
Maria Thor 1,4, Kelly Fitzgerald 2,4, Aditya Apte 1, Jung Hun Oh 1, Aditi Iyer 1, Otasowie Odiase 2, Saad Nadeem 1, Ellen D. Yorke 1, Jamie Chaft 3, Abraham J. Wu 2, Michael Offin 3, Charles B Simone II 2, Isabel Preeshagul 3, Daphna Y. Gelblum 2, Daniel Gomez 2, Joseph O. Deasy 1, Andreas Rimner 2
1Department of Medical Physics, Memorial Sloan Kettering Cancer Center
2Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center
3Department of Medicine, Memorial Sloan Kettering Cancer Center
https://doi.org/10.1016/j.radonc.2023.109983
This paper examines the critical issue of identifying patients at risk of disease progression after SBRT treatment. The occurrence of disease progression in a significant percentage of cases highlights the necessity for improved predictive tools. In this context, the study utilizes an innovative approach by integrating spiculation as a crucial radiomics feature in the analysis. Spiculation, a visually apparent pattern in imaging, has gained recognition for its potential as a prognostic indicator. By utilizing spiculation alongside other radiomics features, this study seeks to improve the precision and dependability of forecasts related to progression-free survival among early-stage NSCLC patients after SBRT. Incorporating spiculation into the radiomics framework is a noteworthy advance toward more personalized and effective therapeutic approaches for this patient cohort.
Highlights
Varian will support my research project entitled “Longitudinal CBCT radiomics analysis for lung cancer radiotherapy response and prognosis prediction” with $230,000 over 2 years. This is the first research grant from Varian to the Department of Radiation Oncology, Sidney Kimmel Medical College at Thomas Jefferson University. This project can potentially impact the clinical practice of lung cancer patients by using standard imaging modalities (CBCT and 4D-CBCT) to provide early prediction of prognosis and toxicity.
We will develop Jefferson Whole Lung CBCT radiomics framework that provides comprehensive lung analysis for early prediction of local control and normal tissue toxicity during lung radiation therapy. This project will provide predictive models for treatment response and prognosis using structural, morphologic, and functional radiomic features acquired during treatment with Varian CBCT imaging, as well as a unique opportunity to combine Varian technology (4D CBCT) with novel image processing techniques to acquire functional images. This project will also include a clinical study of the novel 4D functional CBCT imaging technique.
I am looking for a highly motivated postdoc to lead this project. If you are interested in working with us, please email me at “Wookjin.Choi@jefferson.edu.”
MICCAI’22 Paper | CMPB’21 Paper | CIRDataset
This library serves as a one-stop solution for analyzing datasets using clinically-interpretable radiomics (CIR) in cancer imaging (https://github.com/choilab-jefferson/CIR). The primary motivation for this comes from our collaborators in radiology and radiation oncology inquiring about the importance of clinically-reported features in state-of-the-art deep learning malignancy/recurrence/treatment response prediction algorithms. Previous methods have performed such prediction tasks but without robust attribution to any clinically reported/actionable features (see extensive literature on the sensitivity of attribution methods to hyperparameters). This motivated us to curate datasets by annotating clinically-reported features at the voxel/vertex level on public datasets (using our published advanced mathematical algorithms) and relating these to prediction tasks (bypassing the “flaky” attribution schemes). With the release of these comprehensively-annotated datasets, we hope that previous malignancy prediction methods can also validate their explanations and provide clinically-actionable insights. We also provide strong end-to-end baselines for extracting these hard-to-compute clinically-reported features and using these in different prediction tasks.
Wookjin Choi1, Navdeep Dahiya2, and Saad Nadeem3
1 Department of Radiation Oncology, Thomas Jefferson University Hospital
2 School of Electrical and Computer Engineering, Georgia Institute of Technology
3 Department of Medical Physics, Memorial Sloan Kettering Cancer Center
Spiculations/lobulations, and sharp/curved spikes on the surface of lung nodules, are good predictors of lung cancer malignancy and hence, are routinely assessed and reported by radiologists as part of the standardized Lung-RADS clinical scoring criteria. Given the 3D geometry of the nodule and 2D slice-by-slice assessment by radiologists, manual spiculation/lobulation annotation is a tedious task and thus no public datasets exist to date for probing the importance of these clinically-reported features in the SOTA malignancy prediction algorithms. As part of this paper, we release a large-scale Clinically-Interpretable Radiomics Dataset, CIRDataset, containing 956 radiologist QA/QC’ed spiculation/lobulation annotations on segmented lung nodules from two public datasets, LIDC-IDRI (N=883) and LUNGx (N=73). We also present an end-to-end deep learning model based on multi-class Voxel2Mesh extension to segment nodules (while preserving spikes), classify spikes (sharp/spiculation and curved/lobulation), and perform malignancy prediction. Previous methods have performed malignancy prediction for LIDC and LUNGx datasets but without robust attribution to any clinically reported/actionable features (due to known hyperparameter sensitivity issues with general attribution schemes). With the release of this comprehensively-annotated dataset and end-to-end deep learning baseline, we hope that malignancy prediction methods can validate their explanations, benchmark against our baseline, and provide clinically-actionable insights. Dataset, code and pre-trained models are available in this repository.
The first CIR dataset, released here, contains almost 1000 radiologist QA/QC’ed spiculation/lobulation annotations (computed using our published LungCancerScreeningRadiomics library and QA/QC’ed by a radiologist) on segmented lung nodules for two public datasets, LIDC (with visual radiologist malignancy RM scores for the entire cohort and pathology-proven malignancy PM labels for a subset) and LUNGx (with pathology-proven size-matched benign/malignant nodules to remove the effect of size on malignancy prediction).
We also release our multi-class Voxel2Mesh extension to provide a strong benchmark for end-to-end deep learning lung nodule segmentation, spikes’ classification (lobulation/spiculation), and malignancy prediction; Voxel2Mesh is the only published method to our knowledge that preserves sharp spikes during segmentation and hence its use as our base model. With the release of this comprehensively-annotated dataset, we hope that previous malignancy prediction methods can also validate their explanations/attributions and provide clinically-actionable insights. Users can also generate spiculation/lobulation annotations from scratch for LIDC/LUNGx as well as new datasets using our LungCancerScreeningRadiomics library.
The following tables show the expected results of running the pre-trained ‘Mesh Only’ and ‘Mesh+Encoder’ models.
Table1. Nodule (Class0), spiculation (Class1), and lobulation (Class2) peak classification metrics
| Training | ||||||
|---|---|---|---|---|---|---|
| Network | Chamfer Weighted Symmetric ↓ | Jaccard Index ↑ | ||||
| Class0 | Class1 | Class2 | Class0 | Class1 | Class2 | |
| Mesh Only | 0.009 | 0.010 | 0.013 | 0.507 | 0.493 | 0.430 |
| Mesh+Encoder | 0.008 | 0.009 | 0.011 | 0.488 | 0.456 | 0.410 |
| Validation | ||||||
| Network | Chamfer Weighted Symmetric ↓ | Jaccard Index ↑ | ||||
| Class0 | Class1 | Class2 | Class0 | Class1 | Class2 | |
| Mesh Only | 0.010 | 0.011 | 0.014 | 0.526 | 0.502 | 0.451 |
| Mesh+Encoder | 0.014 | 0.015 | 0.018 | 0.488 | 0.472 | 0.433 |
| Testing LIDC-PM N=72 | ||||||
| Network | Chamfer Weighted Symmetric ↓ | Jaccard Index ↑ | ||||
| Class0 | Class1 | Class2 | Class0 | Class1 | Class2 | |
| Mesh Only | 0.011 | 0.011 | 0.014 | 0.561 | 0.553 | 0.510 |
| Mesh+Encoder | 0.009 | 0.010 | 0.012 | 0.558 | 0.541 | 0.507 |
| Testing LUNGx N=73 | ||||||
| Network | Chamfer Weighted Symmetric ↓ | Jaccard Index ↑ | ||||
| Class0 | Class1 | Class2 | Class0 | Class1 | Class2 | |
| Mesh Only | 0.029 | 0.028 | 0.030 | 0.502 | 0.537 | 0.545 |
| Mesh+Encoder | 0.017 | 0.017 | 0.019 | 0.506 | 0.523 | 0.525 |
Table 2. Malignancy prediction metrics.
| Training | |||||
|---|---|---|---|---|---|
| Network | AUC | Accuracy | Sensitivity | Specificity | F1 |
| Mesh Only | 0.885 | 80.25 | 54.84 | 93.04 | 65.03 |
| Mesh+Encoder | 0.899 | 80.71 | 55.76 | 93.27 | 65.94 |
| Validation | |||||
| Network | AUC | Accuracy | Sensitivity | Specificity | F1 |
| Mesh Only | 0.881 | 80.37 | 53.06 | 92.11 | 61.90 |
| Mesh+Encoder | 0.808 | 75.46 | 42.86 | 89.47 | 51.22 |
| Testing LIDC-PM N=72 | |||||
| Network | AUC | Accuracy | Sensitivity | Specificity | F1 |
| Mesh Only | 0.790 | 70.83 | 56.10 | 90.32 | 68.66 |
| Mesh+Encoder | 0.813 | 79.17 | 70.73 | 90.32 | 79.45 |
| Testing LUNGx N=73 | |||||
| Network | AUC | Accuracy | Sensitivity | Specificity | F1 |
| Mesh Only | 0.733 | 68.49 | 80.56 | 56.76 | 71.60 |
| Mesh+Encoder | 0.743 | 65.75 | 86.11 | 45.95 | 71.26 |
A comprehensive framework for lung cancer screening radiomics using LIDC-IDRI and LUNGx dataset.
Source code https://github.com/choilab-jefferson/LungCancerScreeningRadiomics
POST-DOCTORAL POSITION, DEPARTMENT OF RADIATION ONCOLOGY: Thomas Jefferson University is now accepting applications for a post-doctoral fellow in the Department of Radiation Oncology with the Choi lab. The post-doctoral position is for developing AI techniques for image-guided radiation therapy and clinical outcome prediction and decision-making using radiomics, deep learning, and other computationally intensive techniques. Trainees must have the opportunity to carry out supervised biomedical research with the primary objective of developing or extending their research skills and knowledge in preparation for an independent research career.
This is an exciting opportunity to work on an emerging research project funded by Sidney Kimmel Cancer Center on the incorporation of medical imaging AI into radiation therapy and clinical care for cancer patients. The research focuses on the development of image analysis tools for cancer imaging, as well as clinically interpretable radiomics features and deep learning models for clinical outcome prediction and decision making. The post-doctoral fellow will have an opportunity to collaborate with Thomas Jefferson faculty, national and international collaborators, and work alongside investigators at the NCI Quantitative Imaging Network.
The entire Thomas Jefferson Medical Physics Division currently consists of 22 physicists and eight physics residents. The Jefferson physics faculty lead a highly-impactful and diverse research program. The group currently has ongoing funded projects by the National Cancer Institute, vendor-funded research, as well as Sidney Kimmel Cancer Center funding. System-wide equipment includes a ViewRay MRI-Linac, Varian TrueBeams, and Elekta Agility Linacs. The department has advanced scripting capabilities with multiple Radiation Oncology software packages including Mosaiq, Eclipse, and MIM Software.
Thomas Jefferson University is an Equal Opportunity Employer. Jefferson values diversity and encourages applications from women, members of minority groups, LGBTQ individuals, disabled individuals, and veterans. Applicants should forward a curriculum vitae and a statement of interest to the administrative assistant, Juli Johnson at julianne.johnson@jefferson.edu.
Candidates must have a Ph.D. in Computer Science, Medical Physics, Physics, Mathematics, Electrical Engineering, Biomedical Engineering, or related field required. The ideal candidate will be highly interested in an academic Medical Physics career, have strong computational skills, and seek out a highly collaborative environment. Based on the interest of the post-doctoral fellow; opportunities will be provided to obtain clinical experience, treatment planning experience, as well as mentorship on clinical trial design and statistical modeling methods.
Covid Vaccination is a requirement for employment at Jefferson for employees working at Jefferson’s clinical entities or at the University. If you are not currently vac
Qualia Radiomics 