AI Body Composition in Lung Cancer Screening: Added Value Beyond Lung Cancer Detection.

Background An artificial intelligence (AI) algorithm has been developed for fully automated body composition assessment of lung cancer screening noncontrast low-dose CT of the chest (LDCT) scans, but the utility of these measurements in disease risk prediction models has not been assessed. Purpose To evaluate the added value of CT-based AI-derived body composition measurements in risk prediction of lung cancer incidence, lung cancer death, cardiovascular disease (CVD) death, and all-cause mortality in the National Lung Screening Trial (NLST). Materials and Methods In this secondary analysis of the NLST, body composition measurements, including area and attenuation attributes of skeletal muscle and subcutaneous adipose tissue, were derived from baseline LDCT examinations by using a previously developed AI algorithm. The added value of these measurements was assessed with sex- and cause-specific Cox proportional hazards models with and without the AI-derived body composition measurements for predicting lung cancer incidence, lung cancer death, CVD death, and all-cause mortality. Models were adjusted for confounding variables including age; body mass index; quantitative emphysema; coronary artery calcification; history of diabetes, heart disease, hypertension, and stroke; and other PLCOM2012 lung cancer risk factors. Goodness-of-fit improvements were assessed with the likelihood ratio test. Results Among 20 768 included participants (median age, 61 years [IQR, 57-65 years]; 12 317 men), 865 were diagnosed with lung cancer and 4180 died during follow-up. Including the AI-derived body composition measurements improved risk prediction for lung cancer death (male participants: χ2 = 23.09, P < .001; female participants: χ2 = 15.04, P = .002), CVD death (males: χ2 = 69.94, P < .001; females: χ2 = 16.60, P < .001), and all-cause mortality (males: χ2 = 248.13, P < .001; females: χ2 = 94.54, P < .001), but not for lung cancer incidence (male participants: χ2 = 2.53, P = .11; female participants: χ2 = 1.73, P = .19). Conclusion The body composition measurements automatically derived from baseline low-dose CT examinations added predictive value for lung cancer death, CVD death, and all-cause death, but not for lung cancer incidence in the NLST. Clinical trial registration no. NCT00047385 © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Fintelmann in this issue.

Curating retrospective multimodal and longitudinal data for community cohorts at risk for lung cancer.

BACKGROUND: Large community cohorts are useful for lung cancer research, allowing for the analysis of risk factors and development of predictive models. OBJECTIVE: A robust methodology for (1) identifying lung cancer and pulmonary nodules diagnoses as well as (2) associating multimodal longitudinal data with these events from electronic health record (EHRs) is needed to optimally curate cohorts at scale. METHODS: In this study, we leveraged (1) SNOMED concepts to develop ICD-based decision rules for building a cohort that captured lung cancer and pulmonary nodules and (2) clinical knowledge to define time windows for collecting longitudinal imaging and clinical concepts. We curated three cohorts with clinical data and repeated imaging for subjects with pulmonary nodules from our Vanderbilt University Medical Center. RESULTS: Our approach achieved an estimated sensitivity 0.930 (95% CI: [0.879, 0.969]), specificity of 0.996 (95% CI: [0.989, 1.00]), positive predictive value of 0.979 (95% CI: [0.959, 1.000]), and negative predictive value of 0.987 (95% CI: [0.976, 0.994]) for distinguishing lung cancer from subjects with SPNs. CONCLUSION: This work represents a general strategy for high-throughput curation of multi-modal longitudinal cohorts at risk for lung cancer from routinely collected EHRs.

Lung CT harmonization of paired reconstruction kernel images using generative adversarial networks.

BACKGROUND: The kernel used in CT image reconstruction is an important factor that determines the texture of the CT image. Consistency of reconstruction kernel choice is important for quantitative CT-based assessment as kernel differences can lead to substantial shifts in measurements unrelated to underlying anatomical structures. PURPOSE: In this study, we investigate kernel harmonization in a multi-vendor low-dose CT lung cancer screening cohort and evaluate our approach's validity in quantitative CT-based assessments. METHODS: Using the National Lung Screening Trial, we identified CT scan pairs of the same sessions with one reconstructed from a soft tissue kernel and one from a hard kernel. In total, 1000 pairs of five different paired kernel types (200 each) were identified. We adopt the pix2pix architecture to train models for kernel conversion. Each model was trained on 100 pairs and evaluated on 100 withheld pairs. A total of 10 models were implemented. We evaluated the efficacy of kernel conversion based on image similarity metrics including root mean squared error (RMSE), peak signal-to-noise ratio (PSNR), and structural similarity index measure (SSIM) as well as the capability of the models to reduce measurement shifts in quantitative emphysema and body composition measurements. Additionally, we study the reproducibility of standard radiomic features for all kernel pairs before and after harmonization. RESULTS: Our approach effectively converts CT images from one kernel to another in all paired kernel types, as indicated by the reduction in RMSE (p < 0.05) and an increase in the PSNR (p < 0.05) and SSIM (p < 0.05) for both directions of conversion for all pair types. In addition, there is an increase in the agreement for percent emphysema, skeletal muscle area, and subcutaneous adipose tissue (SAT) area for both directions of conversion. Furthermore, radiomic features were reproducible when compared with the ground truth features. CONCLUSIONS: Kernel conversion using deep learning reduces measurement variation in percent emphysema, muscle area, and SAT area.

The Lung Cancer Prediction Model "Stress Test": Assessment of Models' Performance in a High-Risk Prospective Pulmonary Nodule Cohort.

BACKGROUND: Pulmonary nodules represent a growing health care burden because of delayed diagnosis of malignant lesions and overtesting for benign processes. Clinical prediction models were developed to inform physician assessment of pretest probability of nodule malignancy but have not been validated in a high-risk cohort of nodules for which biopsy was ultimately performed. RESEARCH QUESTION: Do guideline-recommended prediction models sufficiently discriminate between benign and malignant nodules when applied to cases referred for biopsy by navigational bronchoscopy? STUDY DESIGN AND METHODS: We assembled a prospective cohort of 322 indeterminate pulmonary nodules in 282 patients referred to a tertiary medical center for diagnostic navigational bronchoscopy between 2017 and 2019. We calculated the probability of malignancy for each nodule using the Brock model, Mayo Clinic model, and Veterans Affairs (VA) model. On a subset of 168 patients who also had PET-CT scans before biopsy, we also calculated the probability of malignancy using the Herder model. The performance of the models was evaluated by calculating the area under the receiver operating characteristic curves (AUCs) for each model. RESULTS: The study cohort contained 185 malignant and 137 benign nodules (57% prevalence of malignancy). The malignant and benign cohorts were similar in terms of size, with a median longest diameter for benign and malignant nodules of 15 and 16 mm, respectively. The Brock model, Mayo Clinic model, and VA model showed similar performance in the entire cohort (Brock AUC, 0.70; 95% CI, 0.64-0.76; Mayo Clinic AUC, 0.70; 95% CI, 0.64-0.76; VA AUC, 0.67; 95% CI, 0.62-0.74). For 168 nodules with available PET-CT scans, the Herder model had an AUC of 0.77 (95% CI, 0.68-0.85). INTERPRETATION: Currently available clinical models provide insufficient discrimination between benign and malignant nodules in the common clinical scenario in which a patient is being referred for biopsy, especially when PET-CT scan information is not available.

Quantifying emphysema in lung screening computed tomography with robust automated lobe segmentation.

Purpose: Anatomy-based quantification of emphysema in a lung screening cohort has the potential to improve lung cancer risk stratification and risk communication. Segmenting lung lobes is an essential step in this analysis, but leading lobe segmentation algorithms have not been validated for lung screening computed tomography (CT). Approach: In this work, we develop an automated approach to lobar emphysema quantification and study its association with lung cancer incidence. We combine self-supervised training with level set regularization and finetuning with radiologist annotations on three datasets to develop a lobe segmentation algorithm that is robust for lung screening CT. Using this algorithm, we extract quantitative CT measures for a cohort (n=1189) from the National Lung Screening Trial and analyze the multivariate association with lung cancer incidence. Results: Our lobe segmentation approach achieved an external validation Dice of 0.93, significantly outperforming a leading algorithm at 0.90 (p<0.01). The percentage of low attenuation volume in the right upper lobe was associated with increased lung cancer incidence (odds ratio: 1.97; 95% CI: [1.06, 3.66]) independent of PLCOm2012 risk factors and diagnosis of whole lung emphysema. Quantitative lobar emphysema improved the goodness-of-fit to lung cancer incidence (χ2=7.48, p=0.02). Conclusions: We are the first to develop and validate an automated lobe segmentation algorithm that is robust to smoking-related pathology. We discover a quantitative risk factor, lending further evidence that regional emphysema is independently associated with increased lung cancer incidence. The algorithm is provided at https://github.com/MASILab/EmphysemaSeg.

Time-distance vision transformers in lung cancer diagnosis from longitudinal computed tomography.

Features learned from single radiologic images are unable to provide information about whether and how much a lesion may be changing over time. Time-dependent features computed from repeated images can capture those changes and help identify malignant lesions by their temporal behavior. However, longitudinal medical imaging presents the unique challenge of sparse, irregular time intervals in data acquisition. While self-attention has been shown to be a versatile and efficient learning mechanism for time series and natural images, its potential for interpreting temporal distance between sparse, irregularly sampled spatial features has not been explored. In this work, we propose two interpretations of a time-distance vision transformer (ViT) by using (1) vector embeddings of continuous time and (2) a temporal emphasis model to scale self-attention weights. The two algorithms are evaluated based on benign versus malignant lung cancer discrimination of synthetic pulmonary nodules and lung screening computed tomography studies from the National Lung Screening Trial (NLST). Experiments evaluating the time-distance ViTs on synthetic nodules show a fundamental improvement in classifying irregularly sampled longitudinal images when compared to standard ViTs. In cross-validation on screening chest CTs from the NLST, our methods (0.785 and 0.786 AUC respectively) significantly outperform a cross-sectional approach (0.734 AUC) and match the discriminative performance of the leading longitudinal medical imaging algorithm (0.779 AUC) on benign versus malignant classification. This work represents the first self-attention-based framework for classifying longitudinal medical images. Our code is available at https://github.com/tom1193/time-distance-transformer.

Characterizing browser-based medical imaging AI with serverless edge computing: towards addressing clinical data security constraints.

Artificial intelligence (AI) has been widely introduced to various medical imaging applications ranging from disease visualization to medical decision support. However, data privacy has become an essential concern in clinical practice of deploying the deep learning algorithms through cloud computing. The sensitivity of patient health information (PHI) commonly limits network transfer, installation of bespoke desktop software, and access to computing resources. Serverless edge-computing shed light on privacy preserved model distribution maintaining both high flexibility (as cloud computing) and security (as local deployment). In this paper, we propose a browser-based, cross-platform, and privacy preserved medical imaging AI deployment system working on consumer-level hardware via serverless edge-computing. Briefly we implement this system by deploying a 3D medical image segmentation model for computed tomography (CT) based lung cancer screening. We further curate tradeoffs in model complexity and data size by characterizing the speed, memory usage, and limitations across various operating systems and browsers. Our implementation achieves a deployment with (1) a 3D convolutional neural network (CNN) on CT volumes (256×256×256 resolution), (2) an average runtime of 80 seconds across Firefox v.102.0.1/Chrome v.103.0.5060.114/Microsoft Edge v.103.0.1264.44 and 210 seconds on Safari v.14.1.1, and (3) an average memory usage of 1.5 GB on Microsoft Windows laptops, Linux workstation, and Apple Mac laptops. In conclusion, this work presents a privacy-preserved solution for medical imaging AI applications that minimizes the risk of PHI exposure. We characterize the tools, architectures, and parameters of our framework to facilitate the translation of modern deep learning methods into routine clinical care.

Age-related Muscle Fat Infiltration in Lung Screening Participants: Impact of Smoking Cessation.

Rationale: Skeletal muscle fat infiltration progresses with aging and is worsened among individuals with a history of cigarette smoking. Many negative impacts of smoking on muscles are likely reversible with smoking cessation. Objectives: To determine if the progression of skeletal muscle fat infiltration with aging is altered by smoking cessation among lung cancer screening participants. Methods: This was a secondary analysis based on the National Lung Screening Trial. Skeletal muscle attenuation in Hounsfield unit (HU) was derived from the baseline and follow-up low-dose CT scans using a previously validated artificial intelligence algorithm. Lower attenuation indicates greater fatty infiltration. Linear mixed-effects models were constructed to evaluate the associations between smoking status and the muscle attenuation trajectory. Measurements and Main Results: Of 19,019 included participants (age: 61 years, 5 [SD]; 11,290 males), 8,971 (47.2%) were actively smoking cigarettes. Accounting for body mass index, pack-years, percent emphysema, and other confounding factors, actively smoking predicted a lower attenuation in both males (ß0 =-0.88 HU, P<.001) and females (ß0 =-0.69 HU, P<.001), and an accelerated muscle attenuation decline-rate in males (ß1=-0.08 HU/y, P<.05). Age-stratified analyses indicated that the accelerated muscle attenuation decline associated with smoking likely occurred at younger age, especially in females. Conclusions: Among lung cancer screening participants, active cigarette smoking was associated with greater skeletal muscle fat infiltration in both males and females, and accelerated muscle adipose accumulation rate in males. These findings support the important role of smoking cessation in preserving muscle health.

Longitudinal Multimodal Transformer Integrating Imaging and Latent Clinical Signatures From Routine EHRs for Pulmonary Nodule Classification.

The accuracy of predictive models for solitary pulmonary nodule (SPN) diagnosis can be greatly increased by incorporating repeat imaging and medical context, such as electronic health records (EHRs). However, clinically routine modalities such as imaging and diagnostic codes can be asynchronous and irregularly sampled over different time scales which are obstacles to longitudinal multimodal learning. In this work, we propose a transformer-based multimodal strategy to integrate repeat imaging with longitudinal clinical signatures from routinely collected EHRs for SPN classification. We perform unsupervised disentanglement of latent clinical signatures and leverage time-distance scaled self-attention to jointly learn from clinical signatures expressions and chest computed tomography (CT) scans. Our classifier is pretrained on 2,668 scans from a public dataset and 1,149 subjects with longitudinal chest CTs, billing codes, medications, and laboratory tests from EHRs of our home institution. Evaluation on 227 subjects with challenging SPNs revealed a significant AUC improvement over a longitudinal multimodal baseline (0.824 vs 0.752 AUC), as well as improvements over a single cross-section multimodal scenario (0.809 AUC) and a longitudinal imaging-only scenario (0.741 AUC). This work demonstrates significant advantages with a novel approach for co-learning longitudinal imaging and non-imaging phenotypes with transformers. Code available at https://github.com/MASILab/lmsignatures.