Ultra-low-dose vs. standard-of-care-dose CT of the chest in patients with post-COVID-19 conditions-a prospective intra-patient multi-reader study.

OBJECTIVES: To conduct an intrapatient comparison of ultra-low-dose computed tomography (ULDCT) and standard-of-care-dose CT (SDCT) of the chest in terms of the diagnostic accuracy of ULDCT and intrareader agreement in patients with post-COVID conditions. METHODS: We prospectively included 153 consecutive patients with post-COVID-19 conditions. All participants received an SDCT and an additional ULDCT scan of the chest. SDCTs were performed with standard imaging parameters and ULDCTs at a fixed tube voltage of 100 kVp (with tin filtration), 50 ref. mAs (dose modulation active), and iterative reconstruction algorithm level 5 of 5. All CT scans were separately evaluated by four radiologists for the presence of lung changes and their consistency with post-COVID lung abnormalities. Radiation dose parameters and the sensitivity, specificity, and accuracy of ULDCT were calculated. RESULTS: Of the 153 included patients (mean age 47.4 ± 15.3 years; 48.4% women), 45 (29.4%) showed post-COVID lung abnormalities. In those 45 patients, the most frequently detected CT patterns were ground-glass opacities (100.0%), reticulations (43.5%), and parenchymal bands (37.0%). The accuracy, sensitivity, and specificity of ULDCT compared to SDCT for the detection of post-COVID lung abnormalities were 92.6, 87.2, and 94.9%, respectively. The median total dose length product (DLP) of ULDCTs was less than one-tenth of the radiation dose of our SDCTs (12.6 mGy*cm [9.9; 15.5] vs. 132.1 mGy*cm [103.9; 160.2]; p < 0.001). CONCLUSION: ULDCT of the chest offers high accuracy in the detection of post-COVID lung abnormalities compared to an SDCT scan at less than one-tenth the radiation dose, corresponding to only twice the dose of a standard chest radiograph in two views. CLINICAL RELEVANCE STATEMENT: Ultra-low-dose CT of the chest may provide a favorable, radiation-saving alternative to standard-dose CT in the long-term follow-up of the large patient cohort of post-COVID-19 patients.

Contrast Agent Dynamics Determine Radiomics Profiles in Oncologic Imaging.

BACKGROUND: The reproducibility of radiomics features extracted from CT and MRI examinations depends on several physiological and technical factors. The aim was to evaluate the impact of contrast agent timing on the stability of radiomics features using dynamic contrast-enhanced perfusion CT (dceCT) or MRI (dceMRI) in prostate and lung cancers. METHODS: Radiomics features were extracted from dceCT or dceMRI images in patients with biopsy-proven peripheral prostate cancer (pzPC) or biopsy-proven non-small cell lung cancer (NSCLC), respectively. Features that showed significant differences between contrast phases were identified using linear mixed models. An L2-penalized logistic regression classifier was used to predict class labels for pzPC and unaffected prostate regions-of-interest (ROIs). RESULTS: Nine pzPC and 28 NSCLC patients, who were imaged with dceCT and/or dceMRI, were included in this study. After normalizing for individual enhancement patterns by defining seven individual phases based on a reference vessel, 19, 467 and 128 out of 1204 CT features showed significant temporal dynamics in healthy prostate parenchyma, prostate tumors and lung tumors, respectively. CT radiomics-based classification accuracy of healthy and tumor ROIs was highly dependent on contrast agent phase. For dceMRI, 899 and 1027 out of 1118 features were significantly dependent on time after contrast agent injection for prostate and lung tumors. CONCLUSIONS: CT and MRI radiomics features in both prostate and lung tumors are significantly affected by interindividual differences in contrast agent dynamics.

Improving Vessel Segmentation with Multi-Task Learning and Auxiliary Data Available Only During Model Training.

Liver vessel segmentation in magnetic resonance imaging data is important for the computational analysis of vascular remodeling, associated with a wide spectrum of diffuse liver diseases. Existing approaches rely on contrast enhanced imaging data, but the necessary dedicated imaging sequences are not uniformly acquired. Images without contrast enhancement are acquired more frequently, but vessel segmentation is challenging, and requires large-scale annotated data. We propose a multi-task learning framework to segment vessels in liver MRI without contrast. It exploits auxiliary contrast enhanced MRI data available only during training to reduce the need for annotated training examples. Our approach draws on paired native and contrast enhanced data with and without vessel annotations for model training. Results show that auxiliary data improves the accuracy of vessel segmentation, even if they are not available during inference. The advantage is most pronounced if only few annotations are available for training, since the feature representation benefits from the shared task structure. A validation of this approach to augment a model for brain tumor segmentation confirms its benefits across different domains. An auxiliary informative imaging modality can augment expert annotations even if it is only available during training.

Induction of Arterial Inflammation by Immune Checkpoint Inhibitor Therapy in Lung Cancer Patients as Measured by 2-[18F]FDG Positron Emission Tomography/Computed Tomography Depends on Pre-Existing Vascular Inflammation.

BACKGROUND: Immune checkpoint inhibitors (ICI) are one of the most effective therapies in oncology, albeit associated with various immune-related adverse events also affecting the cardiovascular system. METHODS: We aimed to investigate the effect of ICI on arterial 2-[18F]FDG uptake by using 2-[18F]FDG PET/CT imaging pre/post treatment in 47 patients with lung cancer. Maximum 2-[18F]FDG standardized uptake values (SUVmax) and target-to-background ratios (TBRs) were calculated along six arterial segments. We classified the arterial PET lesions by pre-existing active inflammation (cut-off: TBRpre ≥ 1.6). 2-[18F]FDG metabolic activity pre/post treatment was also quantified in bone marrow, spleen, and liver. Circulating blood biomarkers were additionally collected at baseline and after immunotherapy. RESULTS: ICI treatment resulted in significantly increased arterial inflammatory activity, detected by increased TBRs, in all arterial PET lesions analyzed. In particular, a significant elevation of arterial 2-[18F]FDG uptake was only recorded in PET lesions without pre-existing inflammation, in calcified as well as in non-calcified lesions. Furthermore, a significant increase in arterial 2-[18F]FDG metabolic activity after immunotherapy was solely observed in patients not previously treated with chemotherapy or radiotherapy as well as in those without CV risk factors. No significant changes were recorded in either 2-[18F]FDG uptake of bone marrow, spleen and liver after treatment, or the blood biomarkers. CONCLUSIONS: ICI induces vascular inflammation in lung cancer patients lacking pre-existing arterial inflammation.

Deep learning for [18F]fluorodeoxyglucose-PET-CT classification in patients with lymphoma: a dual-centre retrospective analysis.

BACKGROUND: The rising global cancer burden has led to an increasing demand for imaging tests such as [18F]fluorodeoxyglucose ([18F]FDG)-PET-CT. To aid imaging specialists in dealing with high scan volumes, we aimed to train a deep learning artificial intelligence algorithm to classify [18F]FDG-PET-CT scans of patients with lymphoma with or without hypermetabolic tumour sites. METHODS: In this retrospective analysis we collected 16 583 [18F]FDG-PET-CTs of 5072 patients with lymphoma who had undergone PET-CT before or after treatment at the Memorial Sloa Kettering Cancer Center, New York, NY, USA. Using maximum intensity projection (MIP), three dimensional (3D) PET, and 3D CT data, our ResNet34-based deep learning model (Lymphoma Artificial Reader System [LARS]) for [18F]FDG-PET-CT binary classification (Deauville 1-3 vs 4-5), was trained on 80% of the dataset, and tested on 20% of this dataset. For external testing, 1000 [18F]FDG-PET-CTs were obtained from a second centre (Medical University of Vienna, Vienna, Austria). Seven model variants were evaluated, including MIP-based LARS-avg (optimised for accuracy) and LARS-max (optimised for sensitivity), and 3D PET-CT-based LARS-ptct. Following expert curation, areas under the curve (AUCs), accuracies, sensitivities, and specificities were calculated. FINDINGS: In the internal test cohort (3325 PET-CTs, 1012 patients), LARS-avg achieved an AUC of 0·949 (95% CI 0·942-0·956), accuracy of 0·890 (0·879-0·901), sensitivity of 0·868 (0·851-0·885), and specificity of 0·913 (0·899-0·925); LARS-max achieved an AUC of 0·949 (0·942-0·956), accuracy of 0·868 (0·858-0·879), sensitivity of 0·909 (0·896-0·924), and specificity of 0·826 (0·808-0·843); and LARS-ptct achieved an AUC of 0·939 (0·930-0·948), accuracy of 0·875 (0·864-0·887), sensitivity of 0·836 (0·817-0·855), and specificity of 0·915 (0·901-0·927). In the external test cohort (1000 PET-CTs, 503 patients), LARS-avg achieved an AUC of 0·953 (0·938-0·966), accuracy of 0·907 (0·888-0·925), sensitivity of 0·874 (0·843-0·904), and specificity of 0·949 (0·921-0·960); LARS-max achieved an AUC of 0·952 (0·937-0·965), accuracy of 0·898 (0·878-0·916), sensitivity of 0·899 (0·871-0·926), and specificity of 0·897 (0·871-0·922); and LARS-ptct achieved an AUC of 0·932 (0·915-0·948), accuracy of 0·870 (0·850-0·891), sensitivity of 0·827 (0·793-0·863), and specificity of 0·913 (0·889-0·937). INTERPRETATION: Deep learning accurately distinguishes between [18F]FDG-PET-CT scans of lymphoma patients with and without hypermetabolic tumour sites. Deep learning might therefore be potentially useful to rule out the presence of metabolically active disease in such patients, or serve as a second reader or decision support tool. FUNDING: National Institutes of Health-National Cancer Institute Cancer Center Support Grant.

Advancing COVID-19 diagnosis with privacy-preserving collaboration in artificial intelligence.

Artificial intelligence provides a promising solution for streamlining COVID-19 diagnoses; however, concerns surrounding security and trustworthiness impede the collection of large-scale representative medical data, posing a considerable challenge for training a well-generalized model in clinical practices. To address this, we launch the Unified CT-COVID AI Diagnostic Initiative (UCADI), where the artificial intelligence (AI) model can be distributedly trained and independently executed at each host institution under a federated learning framework without data sharing. Here we show that our federated learning framework model considerably outperformed all of the local models (with a test sensitivity/specificity of 0.973/0.951 in China and 0.730/0.942 in the United Kingdom), achieving comparable performance with a panel of professional radiologists. We further evaluated the model on the hold-out (collected from another two hospitals without the federated learning framework) and heterogeneous (acquired with contrast materials) data, provided visual explanations for decisions made by the model, and analysed the trade-offs between the model performance and the communication costs in the federated training process. Our study is based on 9,573 chest computed tomography scans from 3,336 patients collected from 23 hospitals located in China and the United Kingdom. Collectively, our work advanced the prospects of utilizing federated learning for privacy-preserving AI in digital health.