Fully automatic reconstruction of prostate high-dose-rate brachytherapy interstitial needles using two-phase deep learning-based segmentation and object tracking algorithms
The critical aspect of successful brachytherapy (BT) is accurate detection of applicator/needle trajectories, which is an ongoing challenge. This study proposes a two-phase deep learning-based method to automate localization of high-dose-rate (HDR) prostate BT catheters through the patient’s CT imag...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-03-01
|
| Series: | Clinical and Translational Radiation Oncology |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2405630825000151 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832582152942780416 |
|---|---|
| author | Mohammad Mahdi Moradi Zahra Siavashpour Soheib Takhtardeshir Eman Showkatian Ramin Jaberi Reza Ghaderi Bahram Mofid Farzad Taghizadeh-Hesary |
| author_facet | Mohammad Mahdi Moradi Zahra Siavashpour Soheib Takhtardeshir Eman Showkatian Ramin Jaberi Reza Ghaderi Bahram Mofid Farzad Taghizadeh-Hesary |
| author_sort | Mohammad Mahdi Moradi |
| collection | DOAJ |
| description | The critical aspect of successful brachytherapy (BT) is accurate detection of applicator/needle trajectories, which is an ongoing challenge. This study proposes a two-phase deep learning-based method to automate localization of high-dose-rate (HDR) prostate BT catheters through the patient’s CT images. The whole process is divided into two phases using two different deep neural networks. First, BT needles segmentation was accomplished through a pix2pix Generative Adversarial Neural network (pix2pix GAN). Second, a Generic Object Tracking Using Regression Networks (GOTURN) was used to predict the needle trajectories. These models were trained and tested on a clinical prostate BT dataset. Among the total 25 patients, 5 patients that consist of 592 slices was dedicated to testing sets, and the rest were used as train/validation set. The total number of needles in these slices of CT images was 8764, of which the employed pix2pix network was able to segment 98.72 % (8652 of total). Dice Similarity Coefficient (DSC) and IoU (Intersection over Union) between the network output and the ground truth were 0.95 and 0.90, respectively. Moreover, the F1-score, recall, and precision results were 0.95, 0.93, and 0.97, respectively. Regarding the location of the shafts, the proposed model has an error of 0.41 mm. The current study proposed a novel methodology to automatically localize and reconstruct the prostate HDR-BT interstitial needles through the 3D CT images. The presented method can be utilized as a computer-aided module in clinical applications to automatically detect and delineate the multi-catheters, potentially enhancing the treatment quality. |
| format | Article |
| id | doaj-art-2de06bed088d41b3af261f48ef5773ef |
| institution | Kabale University |
| issn | 2405-6308 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Clinical and Translational Radiation Oncology |
| spelling | doaj-art-2de06bed088d41b3af261f48ef5773ef2025-01-30T05:14:32ZengElsevierClinical and Translational Radiation Oncology2405-63082025-03-0151100925Fully automatic reconstruction of prostate high-dose-rate brachytherapy interstitial needles using two-phase deep learning-based segmentation and object tracking algorithmsMohammad Mahdi Moradi0Zahra Siavashpour1Soheib Takhtardeshir2Eman Showkatian3Ramin Jaberi4Reza Ghaderi5Bahram Mofid6Farzad Taghizadeh-Hesary7Faculty of Electrical Engineering Shahid Beheshti University Tehran IranRadiotherapy Oncology Department Shohada-e Tajrish Educational Hospital Shahid Beheshti University of Medical Science Tehran Iran; Clinical Research and Development Unit of Shohaday Tajrish Hospital Tehran Iran; Corresponding authors at: Radiotherapy Oncology Department, Shohada-e Tajrish Educational Hospital, Shahid Beheshti University of Medical Science, Tehran, Iran (Z. Siavashpour). ENT and Head and Neck Research Center and Department, The Five Senses Health Institute, School of Medicine, Iran University of Medical Sciences, Tehran, Iran (F. Taghizadeh-Hesary).Faculty of Electrical Engineering Shahid Beheshti University Tehran IranFaculty of Medical Sciences Department of Medical Physics Iran University of Medical Science Tehran IranRadiation Oncology Department Yas Hospital Tehran University of Medical Sciences Tehran Iran; Department of Physics University of Surrey Guildford UKENT and Head and Neck Research Center and Department The Five Senses Health Institute School of Medicine Iran University of Medical Sciences Tehran IranRadiotherapy Oncology Department Shohada-e Tajrish Educational Hospital Shahid Beheshti University of Medical Science Tehran IranENT and Head and Neck Research Center and Department The Five Senses Health Institute School of Medicine Iran University of Medical Sciences Tehran Iran; Radiation Oncology Department Iran University of Medical Sciences Tehran Iran; Corresponding authors at: Radiotherapy Oncology Department, Shohada-e Tajrish Educational Hospital, Shahid Beheshti University of Medical Science, Tehran, Iran (Z. Siavashpour). ENT and Head and Neck Research Center and Department, The Five Senses Health Institute, School of Medicine, Iran University of Medical Sciences, Tehran, Iran (F. Taghizadeh-Hesary).The critical aspect of successful brachytherapy (BT) is accurate detection of applicator/needle trajectories, which is an ongoing challenge. This study proposes a two-phase deep learning-based method to automate localization of high-dose-rate (HDR) prostate BT catheters through the patient’s CT images. The whole process is divided into two phases using two different deep neural networks. First, BT needles segmentation was accomplished through a pix2pix Generative Adversarial Neural network (pix2pix GAN). Second, a Generic Object Tracking Using Regression Networks (GOTURN) was used to predict the needle trajectories. These models were trained and tested on a clinical prostate BT dataset. Among the total 25 patients, 5 patients that consist of 592 slices was dedicated to testing sets, and the rest were used as train/validation set. The total number of needles in these slices of CT images was 8764, of which the employed pix2pix network was able to segment 98.72 % (8652 of total). Dice Similarity Coefficient (DSC) and IoU (Intersection over Union) between the network output and the ground truth were 0.95 and 0.90, respectively. Moreover, the F1-score, recall, and precision results were 0.95, 0.93, and 0.97, respectively. Regarding the location of the shafts, the proposed model has an error of 0.41 mm. The current study proposed a novel methodology to automatically localize and reconstruct the prostate HDR-BT interstitial needles through the 3D CT images. The presented method can be utilized as a computer-aided module in clinical applications to automatically detect and delineate the multi-catheters, potentially enhancing the treatment quality.http://www.sciencedirect.com/science/article/pii/S2405630825000151BrachytherapyDeep LearningProstateCathetersNeural Networks |
| spellingShingle | Mohammad Mahdi Moradi Zahra Siavashpour Soheib Takhtardeshir Eman Showkatian Ramin Jaberi Reza Ghaderi Bahram Mofid Farzad Taghizadeh-Hesary Fully automatic reconstruction of prostate high-dose-rate brachytherapy interstitial needles using two-phase deep learning-based segmentation and object tracking algorithms Clinical and Translational Radiation Oncology Brachytherapy Deep Learning Prostate Catheters Neural Networks |
| title | Fully automatic reconstruction of prostate high-dose-rate brachytherapy interstitial needles using two-phase deep learning-based segmentation and object tracking algorithms |
| title_full | Fully automatic reconstruction of prostate high-dose-rate brachytherapy interstitial needles using two-phase deep learning-based segmentation and object tracking algorithms |
| title_fullStr | Fully automatic reconstruction of prostate high-dose-rate brachytherapy interstitial needles using two-phase deep learning-based segmentation and object tracking algorithms |
| title_full_unstemmed | Fully automatic reconstruction of prostate high-dose-rate brachytherapy interstitial needles using two-phase deep learning-based segmentation and object tracking algorithms |
| title_short | Fully automatic reconstruction of prostate high-dose-rate brachytherapy interstitial needles using two-phase deep learning-based segmentation and object tracking algorithms |
| title_sort | fully automatic reconstruction of prostate high dose rate brachytherapy interstitial needles using two phase deep learning based segmentation and object tracking algorithms |
| topic | Brachytherapy Deep Learning Prostate Catheters Neural Networks |
| url | http://www.sciencedirect.com/science/article/pii/S2405630825000151 |
| work_keys_str_mv | AT mohammadmahdimoradi fullyautomaticreconstructionofprostatehighdoseratebrachytherapyinterstitialneedlesusingtwophasedeeplearningbasedsegmentationandobjecttrackingalgorithms AT zahrasiavashpour fullyautomaticreconstructionofprostatehighdoseratebrachytherapyinterstitialneedlesusingtwophasedeeplearningbasedsegmentationandobjecttrackingalgorithms AT soheibtakhtardeshir fullyautomaticreconstructionofprostatehighdoseratebrachytherapyinterstitialneedlesusingtwophasedeeplearningbasedsegmentationandobjecttrackingalgorithms AT emanshowkatian fullyautomaticreconstructionofprostatehighdoseratebrachytherapyinterstitialneedlesusingtwophasedeeplearningbasedsegmentationandobjecttrackingalgorithms AT raminjaberi fullyautomaticreconstructionofprostatehighdoseratebrachytherapyinterstitialneedlesusingtwophasedeeplearningbasedsegmentationandobjecttrackingalgorithms AT rezaghaderi fullyautomaticreconstructionofprostatehighdoseratebrachytherapyinterstitialneedlesusingtwophasedeeplearningbasedsegmentationandobjecttrackingalgorithms AT bahrammofid fullyautomaticreconstructionofprostatehighdoseratebrachytherapyinterstitialneedlesusingtwophasedeeplearningbasedsegmentationandobjecttrackingalgorithms AT farzadtaghizadehhesary fullyautomaticreconstructionofprostatehighdoseratebrachytherapyinterstitialneedlesusingtwophasedeeplearningbasedsegmentationandobjecttrackingalgorithms |