Domain generalization for prostate segmentation in transrectal ultrasound images: A multi-center study

Vesal, Sulaiman; Gayo, Iani; Bhattacharya, Indrani; Natarajan, Shyam; Marks, Leonard S.; Barratt, Dean C.; Fan, Richard E.; Hu, Yipeng; Sonn, Geoffrey A.; Rusu, Mirabela

doi:10.1016/j.media.2022.102620

Cited by 23 publications

(10 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…9 A recent large multi-center study that implemented a 2.5D deep neural network (coordinated dilated residual UNet) for prostate segmentation from TRUS images prior to biopsy reported a Dice coefficient of 0.94. 10 Finally, a study that implemented a shape model-based deep learning method for prostate segmentation from TRUS images, some of which had implanted needles or prostate brachytherapy seeds covering part of the prostate gland, reported a Dice coefficient of 0.88. 11 Our AI algorithm performed quite well, given that a median Dice coefficient of 0.92 was achieved with implanted needle artifacts covering the entire prostate in all cases.…”

Section: Discussionmentioning

confidence: 99%

“…Another study that implemented U‐Net/U‐Net++ architectures for prostate segmentation from TRUS images, some of which had anatomic artifacts, reported Dice coefficient of 0.94 for both architectures 9 . A recent large multi‐center study that implemented a 2.5D deep neural network (coordinated dilated residual UNet) for prostate segmentation from TRUS images prior to biopsy reported a Dice coefficient of 0.94 10 . Finally, a study that implemented a shape model‐based deep learning method for prostate segmentation from TRUS images, some of which had implanted needles or prostate brachytherapy seeds covering part of the prostate gland, reported a Dice coefficient of 0.88 11 .…”

Section: Discussionmentioning

confidence: 99%

“…Dice coefficients ≥ 85% have been achieved in numerous retrospective studies, 8,9 including one involving multiple institutions. 10 Though one study did apply deep-learning to TRUS images, with a subset of images containing implanted needles/seeds covering a part of the prostate, 11 we are not aware of any studies strictly focused on TRUS images of prostates fully implanted by needles. This pilot study aims to evaluate a deeplearning algorithm for contouring the prostate on TRUS images with needles in-place for HDR brachytherapy.…”

Section: Introductionmentioning

confidence: 99%

“…Recently, deep‐learning algorithms have been shown to generate high quality contours of the prostate under TRUS without implanted needles. Dice coefficients ≥ 85% have been achieved in numerous retrospective studies, 8,9 including one involving multiple institutions 10 . Though one study did apply deep‐learning to TRUS images, with a subset of images containing implanted needles/seeds covering a part of the prostate, 11 we are not aware of any studies strictly focused on TRUS images of prostates fully implanted by needles.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Observer preference of artificial intelligence‐generated versus clinical prostate contours for ultrasound‐based high dose rate brachytherapy

King,

Kehayias,

Chaunzwa

et al. 2023

Medical Physics

View full text Add to dashboard Cite

BackgroundFor trans‐rectal ultrasound (TRUS)‐based high dose rate (HDR) prostate brachytherapy, prostate contouring can be challenging due to artifacts from implanted needles, bleeding, and calcifications.PurposeTo evaluate the geometric accuracy and observer preference of an artificial intelligence (AI) algorithm for generating prostate contours on TRUS images with implanted needles.MethodsWe conducted a retrospective study of 150 patients, who underwent HDR brachytherapy. These patients were randomly divided into training (104), validation (26) and testing (20) sets. An AI algorithm was trained/validated utilizing the TRUS image and reference (clinical) contours. The algorithm then provided contours for the test set. For evaluation, we calculated the Dice coefficient between AI and reference prostate contours. We then presented AI and reference contours to eight clinician observers, and asked observers to select their preference. Observers were blinded to the source of contours. We calculated the percentage of cases in which observers preferred AI contours. Lastly, we evaluate whether the presence of AI contours improved the geometric accuracy of prostate contours provided by five resident observers for a 10‐patient subset.ResultsThe median Dice coefficient between AI and reference contours was 0.92 (IQR: 0.90–0.94). Observers preferred AI contours for a median of 57.5% (IQR: 47.5, 65.0) of the test cases. For resident observers, the presence of AI contours was associated with a 0.107 (95% CI: 0.086, 0.128; p < 0.001) improvement in Dice coefficient for the 10‐patient subset.ConclusionThe AI algorithm provided high‐quality prostate contours on TRUS with implanted needles. Further prospective study is needed to better understand how to incorporate AI prostate contours into the TRUS‐based HDR brachytherapy workflow.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Observer preference of artificial intelligence‐generated versus clinical prostate contours for ultrasound‐based high dose rate brachytherapy

King,

Kehayias,

Chaunzwa

et al. 2023

Medical Physics

View full text Add to dashboard Cite

show abstract

“…In consideration of these characteristics, several multicenter studies have focused on generalizing deep learning networks. 13 14 15 These studies have demonstrated the potential applicability of deep learning networks in clinical practice.…”

Section: Introductionmentioning

confidence: 88%

Deep learning-based automatic segmentation of the mandibular canal on panoramic radiographs: A multi-device study

Aung,

Lim,

Han

et al. 2024

Imaging Sci Dent

View full text Add to dashboard Cite

Purpose The objective of this study was to propose a deep-learning model for the detection of the mandibular canal on dental panoramic radiographs. Materials and Methods A total of 2,100 panoramic radiographs (PANs) were collected from 3 different machines: RAYSCAN Alpha (n=700, PAN A), OP-100 (n=700, PAN B), and CS8100 (n=700, PAN C). Initially, an oral and maxillofacial radiologist coarsely annotated the mandibular canals. For deep learning analysis, convolutional neural networks (CNNs) utilizing U-Net architecture were employed for automated canal segmentation. Seven independent networks were trained using training sets representing all possible combinations of the 3 groups. These networks were then assessed using a hold-out test dataset. Results Among the 7 networks evaluated, the network trained with all 3 available groups achieved an average precision of 90.6%, a recall of 87.4%, and a Dice similarity coefficient (DSC) of 88.9%. The 3 networks trained using each of the 3 possible 2-group combinations also demonstrated reliable performance for mandibular canal segmentation, as follows: 1) PAN A and B exhibited a mean DSC of 87.9%, 2) PAN A and C displayed a mean DSC of 87.8%, and 3) PAN B and C demonstrated a mean DSC of 88.4%. Conclusion This multi-device study indicated that the examined CNN-based deep learning approach can achieve excellent canal segmentation performance, with a DSC exceeding 88%. Furthermore, the study highlighted the importance of considering the characteristics of panoramic radiographs when developing a robust deep-learning network, rather than depending solely on the size of the dataset.

show abstract

Disrupting prostate cancer research: Challenge accepted; report from the 2023 Coffey‐Holden Prostate Cancer Academy Meeting

Miyahira,

Kamran,

Jamaspishvili

et al. 2024

The Prostate

View full text Add to dashboard Cite

IntroductionThe 2023 Coffey‐Holden Prostate Cancer Academy (CHPCA) Meeting, themed “Disrupting Prostate Cancer Research: Challenge Accepted,” was convened at the University of California, Los Angeles, Luskin Conference Center, in Los Angeles, CA, from June 22 to 25, 2023.MethodsThe 2023 marked the 10th Annual CHPCA Meeting, a discussion‐oriented scientific think‐tank conference convened annually by the Prostate Cancer Foundation, which centers on innovative and emerging research topics deemed pivotal for advancing critical unmet needs in prostate cancer research and clinical care. The 2023 CHPCA Meeting was attended by 81 academic investigators and included 40 talks across 8 sessions.ResultsThe central topic areas covered at the meeting included: targeting transcription factor neo‐enhancesomes in cancer, AR as a pro‐differentiation and oncogenic transcription factor, why few are cured with androgen deprivation therapy and how to change dogma to cure metastatic prostate cancer without castration, reducing prostate cancer morbidity and mortality with genetics, opportunities for radiation to enhance therapeutic benefit in oligometastatic prostate cancer, novel immunotherapeutic approaches, and the new era of artificial intelligence‐driven precision medicine.DiscussionThis article provides an overview of the scientific presentations delivered at the 2023 CHPCA Meeting, such that this knowledge can help in facilitating the advancement of prostate cancer research worldwide.

show abstract

Domain generalization for prostate segmentation in transrectal ultrasound images: A multi-center study

Cited by 23 publications

References 49 publications

Observer preference of artificial intelligence‐generated versus clinical prostate contours for ultrasound‐based high dose rate brachytherapy

Observer preference of artificial intelligence‐generated versus clinical prostate contours for ultrasound‐based high dose rate brachytherapy

Deep learning-based automatic segmentation of the mandibular canal on panoramic radiographs: A multi-device study

Disrupting prostate cancer research: Challenge accepted; report from the 2023 Coffey‐Holden Prostate Cancer Academy Meeting

Contact Info

Product

Resources

About