The RSNA Pulmonary Embolism CT Dataset

Colak, Errol; Kitamura, Felipe; Hobbs, Stephen B.; Wu, Carol C.; Lungren, Matthew P.; Prevedello, Luciano M.; Kalpathy–Cramer, Jayashree; Ball, Robyn L.; Shih, George; Stein, Anouk; Halabi, Safwan; Altınmakas, Emre; Law, Meng; Kumar, Parveen; Manzalawi, Karam A.; D, Rubio; Sechrist, Jacob W.; Germaine, Pauline; Lopez, Eva Castro; Amerio, Tomas; Gupta, Pushpender; Jain, Manoj; Kay, Fernando Uliana; Lin, Cheng Ting; Sen, Saugata; Revels, Jonathan; Brussaard, Carola; Mongan, John; Abdala, Nitamar; Bearce, Benjamin; Carrete, Henrique; Doğan, Hakan; Huang, Shih-Cheng; Crivellaro, Priscila; Dincler, Seval; Kavnoudias, Helen; Lee, Robin; Lin, Hon‐Yi; Salehinejad, Hojjat; Samorodova, Oleksandra; Santos, Ernandez Rodrigues dos; Seah, Jarrel; Zia, Adil; Arteaga, Veronica; Batra, Kiran; Atzingen, Augusto Castelli von; Chacko, Anith; DiDomenico, Paul; Gill, Ritu R.; Hafez, Mona Ahmed Fouad; John, Susan; Karl, Robert L; Kanne, Jeffrey P.; Nair, Rajesh V. Mathilakath; McDermott, Shaunagh; Mittal, Pardeep; Mumbower, Amy L; Lee, Christopher; Orausclio, Paola; Palacio, Diana; Pozzessere, Chiara; Rajiah, Prabhakar; Ramos, Oswaldo A.; Rodriguez, Sonia; Shaaban, Mahmoud N.; Shah, Palmi; Son, Hong-Ju; Sonavane, Sushilkumar K.; Spieler, Bradley; Tsai, Emily B.; Vasquez, Adriana; Vijayakumar, D.; Wali, Praveen; Wand, Austin; Endara, G. Elizabeth Zamora

doi:10.1148/ryai.2021200254

Cited by 59 publications

(41 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The patients were referred for assessment, follow-up, and roll-out of COVID-19. Besides, three online databases of TCIA chest CT (D1, 600 cases) [ 18 , 19 ], Russia chest CT for assessment of COVID-19 (D2, 1022 cases) [ 20 ], and RSNA chest CT images (D3, 7052 cases) [ 21 ] were included. D1 and D3 databases were created from multiple pathologies.…”

Section: Methodsmentioning

confidence: 99%

Deep learning-based fully automated Z-axis coverage range definition from scout scans to eliminate overscanning in chest CT imaging

et al. 2021

View full text Add to dashboard Cite

Background Despite the prevalence of chest CT in the clinic, concerns about unoptimized protocols delivering high radiation doses to patients still remain. This study aimed to assess the additional radiation dose associated with overscanning in chest CT and to develop an automated deep learning-assisted scan range selection technique to reduce radiation dose to patients. Results A significant overscanning range (31 ± 24) mm was observed in clinical setting for over 95% of the cases. The average Dice coefficient for lung segmentation was 0.96 and 0.97 for anterior–posterior (AP) and lateral projections, respectively. By considering the exact lung coverage as the ground truth, and AP and lateral projections as input, The DL-based approach resulted in errors of 0.08 ± 1.46 and − 1.5 ± 4.1 mm in superior and inferior directions, respectively. In contrast, the error on external scout views was − 0.7 ± 4.08 and 0.01 ± 14.97 mm for superior and inferior directions, respectively.The ED reduction achieved by automated scan range selection was 21% in the test group. The evaluation of a large multi-centric chest CT dataset revealed unnecessary ED of more than 2 mSv per scan and 67% increase in the thyroid absorbed dose. Conclusion The proposed DL-based solution outperformed previous automatic methods with acceptable accuracy, even in complicated and challenging cases. The generizability of the model was demonstrated by fine-tuning the model on AP scout views and achieving acceptable results. The method can reduce the unoptimized dose to patients by exclunding unnecessary organs from field of view.

show abstract

Section: Methodsmentioning

confidence: 99%

Deep learning-based fully automated Z-axis coverage range definition from scout scans to eliminate overscanning in chest CT imaging

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Our public dataset consisted of 5 separate datasets, including 1,744 COVID-19 34,45,46 , pulmonary embolism (PE, n=5,696) 47 , and lung cancer (n=1,379) 45 CT images. Our private datasets were assembled in this effort from 18 clinical centers in Iran totaling 13,404 COVID-19 patients with the same number of CT images (one image per patient).…”

Section: Datasetsmentioning

confidence: 99%

“…In our local medical centers, CT imaging was performed at the end inhalation breath-hold to decrease respiratory motion artefacts. More information on public database were provided in 34,[45][46][47] .…”

Section: Datasetsmentioning

confidence: 99%

Diagnosis of COVID-19 Using CT image Radiomics Features: A Comprehensive Machine Learning Study Involving 26,307 Patients

Shiri

Salimi

Saberi

et al. 2021

Preprint

View full text Add to dashboard Cite

Purpose: To derive and validate an effective radiomics-based model for differentiation of COVID-19 pneumonia from other lung diseases using a very large cohort of patients. Methods: We collected 19 private and 5 public datasets, accumulating to 26,307 individual patient images (15,148 COVID-19; 9,657 with other lung diseases e.g. non-COVID-19 pneumonia, lung cancer, pulmonary embolism; 1502 normal cases). Images were automatically segmented using a validated deep learning (DL) model and the results carefully reviewed. Images were first cropped into lung-only region boxes, then resized to 296 by 216 voxels. Voxel dimensions was resized to 1mm3 followed by 64-bin discretization. The 108 extracted features included shape, first-order histogram and texture features. Univariate analysis was first performed using simple logistic regression. The thresholds were fixed in the training set and then evaluation performed on the test set. False discovery rate (FDR) correction was applied to the p-values. Z-Score normalization was applied to all features. For multivariate analysis, features with high correlation (R2>0.99) were eliminated first using Pearson correlation. We tested 96 different machine learning strategies through cross-combining 4 feature selectors or 8 dimensionality reduction techniques with 8 classifiers. We trained and evaluated our models using 3 different datasets: 1) the entire dataset (26,307 patients: 15,148 COVID-19; 11,159 non-COVID-19); 2) excluding normal patients in non-COVID-19, and including only RT-PCR positive COVID-19 cases in the COVID-19 class (20,697 patients including 12,419 COVID-19, and 8,278 non-COVID-19)); 3) including only non-COVID-19 pneumonia patients and a random sample of COVID-19 patients (5,582 patients: 3,000 COVID-19, and 2,582 non-COVID-19) to provide balanced classes. Subsequently, each of these 3 datasets were randomly split into 70% and 30% for training and testing, respectively. All various steps, including feature preprocessing, feature selection, and classification, were performed separately in each dataset. Classification algorithms were optimized during training using grid search algorithms. The best models were chosen by a one-standard-deviation rule in 10-fold cross-validation and then were evaluated on the test sets. Results: In dataset #1, Relief feature selection and RF classifier combination resulted in the highest performance (Area under the receiver operating characteristic curve (AUC) = 0.99, sensitivity = 0.98, specificity = 0.94, accuracy = 0.96, positive predictive value (PPV) = 0.96, and negative predicted value (NPV) = 0.96). In dataset #2, Recursive Feature Elimination (RFE) feature selection and Random Forest (RF) classifier combination resulted in the highest performance (AUC = 0.99, sensitivity = 0.98, specificity = 0.95, accuracy = 0.97, PPV = 0.96, and NPV = 0.98). In dataset #3, the ANOVA feature selection and RF classifier combination resulted in the highest performance (AUC = 0.98, sensitivity = 0.96, specificity = 0.93, accuracy = 0.94, PPV = 0.93, NPV = 0.96). Conclusion: Radiomic features extracted from entire lung combined with machine learning algorithms can enable very effective, routine diagnosis of COVID-19 pneumonia from CT images without the use of any other diagnostic test.

show abstract

“…Some recent publicly released radiology datasets are quite large. For example, the RSNA CT pulmonary angiography dataset is exceptionally large (20). There is a cost to making very large datasets publicly available, which is still an open issue.…”

Section: Rmsmentioning

confidence: 99%

RSNA-MICCAI Panel Discussion: 2. Leveraging the Full Potential of AI—Radiologists and Data Scientists Working Together

Linguraru¹,

Maier-Hein²,

Summers³

et al. 2021

Radiology: Artificial Intelligence

View full text Add to dashboard Cite

T his article presents the second in a series of panel discussions hosted alternately by Radiological Society of North America (RSNA) and the Medical Image Computing and Computer Assisted Intervention (MICCAI) Society. Both organizations share a vision to advance the science of medical imaging using advanced quantitative imaging biomarkers and artificial intelligence (AI). At the first panel, RSNA members presented virtually as part of the MICCAI 2020 conference (1). The second panel, hosted online by RSNA in March 2021 and reported here, featured three MICCAI members (M.G.L, L.M.H., and R.M.S.). The panelists addressed how radiologists and data scientists can work together to advance the science of AI in radiology.Ranking algorithms in competitions is important, but many algorithms are developed using independent datasets, and comparing algorithms in the literature is quite difficult with the sparse reporting of different metrics. What steps would you suggest for researchers publishing algorithms that are doing so outside of competitions? L.M.H.:The first challenge is to pick a metric or a set of metrics that reflects the clinical goal. We should use standardized metrics for a given class of problems, but there is increasing evidence that some of the commonly used metrics fail to reflect those aspects that matter clinically. This failure is especially important as researchers tend to tune their algorithms according to the performance met-This copy is for personal use only. To order printed copies, contact

show abstract

The RSNA Pulmonary Embolism CT Dataset

Cited by 59 publications

References 11 publications

Deep learning-based fully automated Z-axis coverage range definition from scout scans to eliminate overscanning in chest CT imaging

Deep learning-based fully automated Z-axis coverage range definition from scout scans to eliminate overscanning in chest CT imaging

Diagnosis of COVID-19 Using CT image Radiomics Features: A Comprehensive Machine Learning Study Involving 26,307 Patients

RSNA-MICCAI Panel Discussion: 2. Leveraging the Full Potential of AI—Radiologists and Data Scientists Working Together

Contact Info

Product

Resources

About