Deep learning based automatic segmentation of organs-at-risk for 0.35 T MRgRT of lung tumors

Radiation Oncology

Research

Marco Riboldi¹  Maria Kawula²  Moritz Rabe²  Sebastian Marschner²  Guillaume Landry²  Christopher Kurz²  Stefanie Corradini²  Marvin F. Ribeiro²  Claus Belka³ 

[1] Department of Medical Physics, Ludwig-Maximilians-Universität München, Garching, Germany;Department of Radiation Oncology, LMU University Hospital, LMU Munich, Munich, Germany;Department of Radiation Oncology, LMU University Hospital, LMU Munich, Munich, Germany;German Cancer Consortium (DKTK), Partner Site Munich, Munich, Germany;Bavarian Cancer Research Center (BZKF), Munich, Germany;

关键词: Deep learning;    Auto-segmentation;    MR-Linac;    MRI-guidance;    Thorax;   

DOI  :  10.1186/s13014-023-02330-4

 received in 2023-04-21, accepted in 2023-08-03,  发布年份 2023

来源: Springer

【 摘 要 】

Background and purposeMagnetic resonance imaging guided radiotherapy (MRgRT) offers treatment plan adaptation to the anatomy of the day. In the current MRgRT workflow, this requires the time consuming and repetitive task of manual delineation of organs-at-risk (OARs), which is also prone to inter- and intra-observer variability. Therefore, deep learning autosegmentation (DLAS) is becoming increasingly attractive. No investigation of its application to OARs in thoracic magnetic resonance images (MRIs) from MRgRT has been done so far. This study aimed to fill this gap.Materials and methods122 planning MRIs from patients treated at a 0.35 T MR-Linac were retrospectively collected. Using an 80/19/23 (training/validation/test) split, individual 3D U-Nets for segmentation of the left lung, right lung, heart, aorta, spinal canal and esophagus were trained. These were compared to the clinically used contours based on Dice similarity coefficient (DSC) and Hausdorff distance (HD). They were also graded on their clinical usability by a radiation oncologist.ResultsMedian DSC was 0.96, 0.96, 0.94, 0.90, 0.88 and 0.78 for left lung, right lung, heart, aorta, spinal canal and esophagus, respectively. Median 95th percentile values of the HD were 3.9, 5.3, 5.8, 3.0, 2.6 and 3.5 mm, respectively. The physician preferred the network generated contours over the clinical contours, deeming 85 out of 129 to not require any correction, 25 immediately usable for treatment planning, 15 requiring minor and 4 requiring major corrections.ConclusionsWe trained 3D U-Nets on clinical MRI planning data which produced accurate delineations in the thoracic region. DLAS contours were preferred over the clinical contours.

【 授权许可】

CC BY   
© BioMed Central Ltd., part of Springer Nature 2023

【 图 表 】

Fig. 2

Fig. 3

Fig. 4

Fig. 2

Fig. 2

Fig. 4

Fig. 5

Fig. 2

Fig. 3

Fig. 1

Fig. 2

Fig. 1

Fig. 3

Fig. 1

Fig. 1

Scheme 1

40854_2023_507_Article_IEq31.gif

Fig. 1

13690_2023_1151_Article_IEq1.gif

Fig. 10

Fig. 2

Fig. 1

40854_2023_507_Article_IEq23.gif

Fig. 2

Fig. 1

Fig. 6

Fig. 1

Fig. 2

Fig. 2

40854_2023_507_Article_IEq17.gif

Fig. 5

Fig. 1

40517_2023_266_Article_IEq6.gif

Fig. 1

40517_2023_266_Article_IEq4.gif

Fig. 1

40517_2023_266_Article_IEq1.gif

Fig. 2

40517_2023_266_Article_IEq68.gif

40517_2023_266_Article_IEq67.gif

40517_2023_266_Article_IEq66.gif

Fig. 1

40517_2023_266_Article_IEq64.gif

40517_2023_266_Article_IEq62.gif

40517_2023_266_Article_IEq61.gif

40517_2023_266_Article_IEq60.gif

40517_2023_266_Article_IEq59.gif

40517_2023_266_Article_IEq58.gif

40517_2023_266_Article_IEq57.gif

40517_2023_266_Article_IEq56.gif

40517_2023_266_Article_IEq55.gif

40517_2023_266_Article_IEq54.gif

40517_2023_266_Article_IEq52.gif

40517_2023_266_Article_IEq50.gif

40517_2023_266_Article_IEq49.gif

40517_2023_266_Article_IEq48.gif

40517_2023_266_Article_IEq19.gif

40517_2023_266_Article_IEq18.gif

Fig. 2

12888_2023_5115_Article_IEq2.gif

40517_2023_266_Article_IEq10.gif

Fig. 2

Fig. 1