DICOM re‐encoding of volumetrically annotated Lung Imaging Database Consortium (LIDC) nodules

Fedorov, Andrey; Hancock, Matthew C.; Clunie, David A.; Brochhausen, Mathias; Bona, Jonathan P.; Kirby, Justin; Freymann, John; Pieper, Steve; Aerts, Hugo J.W.L.; Kikinis, Ron; Prior, Fred

doi:10.1002/mp.14445

Cited by 9 publications

(10 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…40,41,44,45,67,68 To that end, DICOM-structured reports (DICOM-SR) were recruited in the ZF GUI/viewer primarily for assigning spatial coordinates and simple shapes linked to coded text labels, all highly applicable to this work (Table 3; Appendix C). 42,43,[69][70][71] In addition, based on the practical experience of the ground-truth expert with the operations of the ZF GUI/viewer to date, noncontributing inference-display redundancy and complexity due to multiple overlapping identically labeled GBBs were reduced (Appendix C).…”

Section: Essential Technical Developments Supporting Real-world Model...mentioning

confidence: 99%

“…To that end, DICOM-structured reports (DICOM-SR) were recruited in the ZF GUI/viewer primarily for assigning spatial coordinates and simple shapes linked to coded text labels. 42,43,[69][70][71] DICOMsegmentation (DICOM-SEG) was also incorporated for future pursuits needing representation of more complex 3D shapes with the flexibility for manual editing during the adjudication process. 42,43,[69][70][71] Based on practical experience of the ground-truth expert with the operations of the ZF GUI/ viewer, noncontributing inference-display redundancy and complexity (i.e., LLIED visualization hindered due to multiple overlapping identically labeled GBBs) was reduced via case-by-case limitation of the stacked inference-GBB display for each identified LLIED type to the one GBB with the highest probability level.…”

Section: Appendix A: Essential Technical Developments Supporting Real...mentioning

confidence: 99%

See 1 more Smart Citation

Pre-deployment assessment of an AI model to assist radiologists in chest X-ray detection and identification of lead-less implanted electronic devices for pre-MRI safety screening: realized implementation needs and proposed operational solutions

et al. 2022

View full text Add to dashboard Cite

Purpose: Chest X-ray (CXR) use in pre-MRI safety screening, such as for lead-less implanted electronic device (LLIED) recognition, is common. To assist CXR interpretation, we "predeployed" an artificial intelligence (AI) model to assess (1) accuracies in LLIED-type (and consequently safety-level) identification, (2) safety implications of LLIED nondetections or misidentifications, (3) infrastructural or workflow requirements, and (4) demands related to model adaptation to real-world conditions.Approach: A two-tier cascading methodology for LLIED detection/localization and identification on a frontal CXR was applied to evaluate the performance of the original nine-class AI model. With the unexpected early appearance of LLIED types during simulated real-world trialing, retraining of a newer 12-class version preceded retrialing. A zero footprint (ZF) graphical user interface (GUI)/viewer with DICOM-based output was developed for inference-result display and adjudication, supporting end-user engagement and model continuous learning and/or modernization.Results: During model testing or trialing using both the nine-class and 12-class models, robust detection/localization was consistently 100%, with mAP 0.99 from fivefold cross-validation. Safety-level categorization was high during both testing (AUC ≥ 0.98 and ≥0.99, respectively) and trialing (accuracy 98% and 97%, respectively). LLIED-type identifications by the two models during testing (1) were 98.9% and 99.5% overall correct and (2) consistently showed AUC ≥ 0.92 (1.00 for 8/9 and 9/12 LLIED-types, respectively). Pre-deployment trialing of both models demonstrated overall type-identification accuracies of 94.5% and 95%, respectively. Of the small number of misidentifications, none involved MRI-stringently conditional or MRIunsafe types of LLIEDs. Optimized ZF GUI/viewer operations led to greater user-friendliness for radiologist engagement.Conclusions: Our LLIED-related AI methodology supports (1) 100% detection sensitivity, (2) high identification (including MRI-safety) accuracy, and (3) future model deployment with facilitated inference-result display and adjudication for ongoing model adaptation to future realworld experiences.

show abstract

Section: Essential Technical Developments Supporting Real-world Model...mentioning

confidence: 99%

Section: Appendix A: Essential Technical Developments Supporting Real...mentioning

confidence: 99%

Pre-deployment assessment of an AI model to assist radiologists in chest X-ray detection and identification of lead-less implanted electronic devices for pre-MRI safety screening: realized implementation needs and proposed operational solutions

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Nodules" by Fedorov, et al 12 In "PleThora: Pleural effusion and thoracic cavity segmentations in diseased lungs for benchmarking chest CT processing pipelines" by Kiser, et al 14 describe a dataset of thoracic cavity segmentations and discrete pleural effusion segmentations annotated on 402 CT scans acquired from patients with non-small cell lung cancer (NSCLC). These data can be used for developing image analysis pipelines such as lung structure segmentation, lesion detection, and radiomics feature extraction.…”

Section: Accepted Articlementioning

confidence: 99%

“…In “DICOM Re‐encoding of Volumetrically Annotated Lung Imaging Data Consortium (LIDC) Nodules” by Fedorov et al 12 describe annotations for lung nodules from 875 of the subjects collected by the Lung Imaging Data Consortium and Image Database Resource Initiative (LIDC) converted into standard DICOM objects to simplify reuse of the data with the readily available open‐source tools, and to improve adherence to FAIR (Findable, Accessible, Interoperable, Reusable) principles 13 …”

Section: Introductionmentioning

confidence: 99%

“…In "DICOM Re-encoding of Volumetrically Annotated Lung Imaging Data Consortium (LIDC) Nodules" by Fedorov et al 12 describe annotations for lung nodules from 875 of the subjects collected by the Lung Imaging Data Consortium and Image Database Resource Initiative (LIDC) converted into standard DICOM objects to simplify reuse of the data with the readily available open-source tools, and to improve adherence to FAIR (Findable, Accessible, Interoperable, Reusable) principles. 13 In "PleThora: Pleural effusion and thoracic cavity segmentations in diseased lungs for benchmarking chest CT processing pipelines" by Kiser et al 14 describe a dataset of thoracic cavity segmentations and discrete pleural effusion segmentations annotated on 402 CT scans acquired from patients with non-small cell lung cancer (NSCLC).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Introduction to special issue on datasets hosted in The Cancer Imaging Archive (TCIA)

et al. 2020

Self Cite

View full text Add to dashboard Cite

Simultaneous object detection and segmentation for patient‐specific markerless lung tumor tracking in simulated radiographs with deep learning

Huang,

Kurz,

Freislederer

et al. 2023

Medical Physics

View full text Add to dashboard Cite

BackgroundReal‐time tumor tracking is one motion management method to address motion‐induced uncertainty. To date, fiducial markers are often required to reliably track lung tumors with X‐ray imaging, which carries risks of complications and leads to prolonged treatment time. A markerless tracking approach is thus desirable. Deep learning‐based approaches have shown promise for markerless tracking, but systematic evaluation and procedures to investigate applicability in individual cases are missing. Moreover, few efforts have been made to provide bounding box prediction and mask segmentation simultaneously, which could allow either rigid or deformable multi‐leaf collimator tracking.PurposeThe purpose of this study was to implement a deep learning‐based markerless lung tumor tracking model exploiting patient‐specific training which outputs both a bounding box and a mask segmentation simultaneously. We also aimed to compare the two kinds of predictions and to implement a specific procedure to understand the feasibility of markerless tracking on individual cases.MethodsWe first trained a Retina U‐Net baseline model on digitally reconstructed radiographs (DRRs) generated from a public dataset containing 875 CT scans and corresponding lung nodule annotations. Afterwards, we used an independent cohort of 97 lung patients to develop a patient‐specific refinement procedure. In order to determine the optimal hyperparameters for automatic patient‐specific training, we selected 13 patients for validation where the baseline model predicted a bounding box on planning CT (PCT)‐DRR with intersection over union (IoU) with the ground‐truth higher than 0.7. The final test set contained the remaining 84 patients with varying PCT‐DRR IoU. For each testing patient, the baseline model was refined on the PCT‐DRR to generate a patient‐specific model, which was then tested on a separate 10‐phase 4DCT‐DRR to mimic the intrafraction motion during treatment. A template matching algorithm served as benchmark model. The testing results were evaluated by four metrics: the center of mass (COM) error and the Dice similarity coefficient (DSC) for segmentation masks, and the center of box (COB) error and the DSC for bounding box detections. Performance was compared to the benchmark model including statistical testing for significance.ResultsA PCT‐DRR IoU value of 0.2 was shown to be the threshold dividing inconsistent (68%) and consistent (100%) success (defined as mean bounding box DSC > 0.6) of PS models on 4DCT‐DRRs. Thirty‐seven out of the eighty‐four testing cases had a PCT‐DRR IoU above 0.2. For these 37 cases, the mean COM error was 2.6 mm, the mean segmentation DSC was 0.78, the mean COB error was 2.7 mm, and the mean box DSC was 0.83. Including the validation cases, the model was applicable to 50 out of 97 patients when using the PCT‐DRR IoU threshold of 0.2. The inference time per frame was 170 ms. The model outperformed the benchmark model on all metrics, and the comparison was significant (p < 0.001) over the 37 PCT‐DRR IoU > 0.2 cases, but not over the undifferentiated 84 testing cases.ConclusionsThe implemented patient‐specific refinement approach based on a pre‐trained baseline model was shown to be applicable to markerless tumor tracking in simulated radiographs for lung cases.

show abstract

DICOM re‐encoding of volumetrically annotated Lung Imaging Database Consortium (LIDC) nodules

Cited by 9 publications

References 36 publications

Pre-deployment assessment of an AI model to assist radiologists in chest X-ray detection and identification of lead-less implanted electronic devices for pre-MRI safety screening: realized implementation needs and proposed operational solutions

Pre-deployment assessment of an AI model to assist radiologists in chest X-ray detection and identification of lead-less implanted electronic devices for pre-MRI safety screening: realized implementation needs and proposed operational solutions

Introduction to special issue on datasets hosted in The Cancer Imaging Archive (TCIA)

Simultaneous object detection and segmentation for patient‐specific markerless lung tumor tracking in simulated radiographs with deep learning

Contact Info

Product

Resources

About