Extraction of vector field-based motion signatures from 4D (CB)CT image sequences and correlation with local control of lung and liver metastases after SBRT

• Applicant: Professor Dr. René Werner
• Subject Area: Epidemiology and Medical Biometry/Statistics ; Medical Physics, Biomedical Technology
• Term: from 2017 to 2024
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 390567362

• Final Report Year: 2024

Final Report Abstract

Respiratory motion represents a challenge in radiotherapy of thoracic and abdominal tumors. To take into account associated uncertainties during radiation planning and therapy, imaging and accurate assessment of tumor motion and motion of organs at risk is required. Corresponding cornerstones of current radiotherapy workflows are 4D computed tomography (4D CT) and 4D cone beam CT (4D CBCT). However, both imaging modalities lead to image artifacts that hamper their clinical application (e.g., target volume definition, non-linear registration-based motion estimation, estimation of the applied dose). Our project aimed to optimize 4D CT and 4D CBCT image acquisition and image quality, image-based extraction of motion information and to better understand the relationship between motion variability, image quality and the most important clinical endpoint: local tumor control. Using deep learning approaches, methods were developed to minimize motion-associated artifacts in 4D CT and 4D CBCT image data as well as errors in registration-based motion field estimation. 4D CT artifacts are mainly caused by motion variability during data acquisition. To tackle this cause directly, a new 4D CT imaging protocol (i4DCT) was developed together with our cooperation partners (Siemens Healthineers). i4DCT adapts the data acquisition in real time to the patient's breathing patterns and leads to a significant artifact reduction compared to conventional 4D CT protocols. This appears even more relevant from a clinical perspective, as a significant correlation was observed between reduced local tumor control and the occurrence of severe artifacts in the 4D CT radiotherapy treatment planning image data. However, it remains to be investigated whether this correlation is causal or whether other aspects are the underlying cause (e.g., differences in breathing patterns during 4D CT imaging and treatment dose application).

Publications

• Intelligent 4D CT sequence scanning (i4DCT): Concept and performance evaluation. Medical Physics, 46(8), 3462-3474.
  Werner, René; Sentker, Thilo; Madesta, Frederic; Gauer, Tobias & Hofmann, Christian
  
• Self-consistent deep learning-based boosting of 4D cone-beam computed tomography reconstruction. Medical Imaging 2019: Image Processing, 1. SPIE.
  Madesta, Frederic; Gauer, Tobias; Sentker, Thilo & Werner, René
  
• 4D CT image artifacts affect local control in SBRT of lung and liver metastases. Radiotherapy and Oncology, 148, 229-234.
  Sentker, Thilo; Schmidt, Vladimir; Ozga, Ann-Kathrin; Petersen, Cordula; Madesta, Frederic; Hofmann, Christian; Werner, René & Gauer, Tobias
  
• Intelligent 4D CT sequence scanning (i4DCT): First scanner prototype implementation and phantom measurements of automated breathing signal‐guided 4D CT. Medical Physics, 47(6), 2408-2412.
  Werner, René; Sentker, Thilo; Madesta, Frederic; Schwarz, Annette; Vornehm, Marc; Gauer, Tobias & Hofmann, Christian
  
• PO-1724: Intelligent 4D CT sequence scanning (i4DCT): First prototype measurements. Radiotherapy and Oncology, 152, S954-S955.
  Werner, R.; Sentker, T.; Madesta, F.; Gauer, T. & Hofmann, C.
  
• Self‐contained deep learning‐based boosting of 4D cone‐beam CT reconstruction. Medical Physics, 47(11), 5619-5631.
  Madesta, Frederic; Sentker, Thilo; Gauer, Tobias & Werner, René
  
• Comparison of intelligent 4D CT sequence scanning and conventional spiral 4D CT: a first comprehensive phantom study. Physics in Medicine & Biology, 66(1), 015004.
  Werner, René; Szkitsak, Juliane; Sentker, Thilo; Madesta, Frederic; Schwarz, Annette; Fernolendt, Susanne; Vornehm, Marc; Gauer, Tobias; Bert, Christoph & Hofmann, Christian
  
• First clinical evaluation of breathing controlled four-dimensional computed tomography imaging. Physics and Imaging in Radiation Oncology, 20, 56-61.
  Szkitsak, Juliane; Werner, René; Fernolendt, Susanne; Schwarz, Annette; Ott, Oliver J.; Fietkau, Rainer; Hofmann, Christian & Bert, Christoph
  
• OC-0560 Impact of 4D-CT imaging protocol on local control in SBRT of lung and liver metastases. Radiotherapy and Oncology, 161, S446.
  Gauer, T.; Sentker, T.; Schmidt, V.; Wimmert, L.; Ozga, A.; Petersen, C.; Madesta, F.; Hofmann, C. & Werner, R.
  
• PH-0655 Filling in the allegedly unknown: Detection and inpainting of interpolation artifacts in 4DCT images. Radiotherapy and Oncology, 161, S525–S526.
  Madesta, F., Sentker, T., Gauer, T. & Werner, R.
  
• OC-0943 Deep learning-based internal target volume adaption in SBRT. Radiotherapy and Oncology, 170, S835-S836.
  Wimmert, L.; Sentker, T.; Dassow, T.; Madesta, F.; Werner, R. & Gauer, T.
  
• Clinical application of breathing-adapted 4D CT: image quality comparison to conventional 4D CT. Strahlentherapie und Onkologie, 199(7), 686-691.
  Werner, René; Szkitsak, Juliane; Madesta, Frederic; Büttgen, Laura; Wimmert, Lukas; Sentker, Thilo; Fietkau, Rainer; Haderlein, Marlen; Bert, Christoph; Gauer, Tobias & Hofmann, Christian
  
• Deep learning‐based conditional inpainting for restoration of artifact‐affected 4D CT images. Medical Physics, 51(5), 3437-3454.
  Madesta, Frederic; Sentker, Thilo; Gauer, Tobias & Werner, René
  
• Dose reduction in sequence scanning 4D CT imaging through respiratory signal‐guided tube current modulation: A feasibility study. Medical Physics, 50(12), 7539-7547.
  Schwarz, Annette; Werner, René; Wimmert, Lukas; Vornehm, Marc; Gauer, Tobias & Hofmann, Christian
  
• PD-0666 Dose reduction for respiratory signal-guided step-and-shoot 4DCT by online dose modulation. Radiotherapy and Oncology, 182, S558–S559.
  Werner, R., Schwarz, A., Wimmert, L., Buttgen, L., Vornehm, M., Gauer, T. & Hofmann, C.
  
• PO-1886 Respiratory motion prediction based on LSTM and linear regression models. Radiotherapy and Oncology, 182, S1629-S1630.
  Wimmert, L.; Nielsen, M.; Gauer, T.; Hofmann, C. & Werner, R.
  
• PO-1917 Stability assessment of patient-specific 4DCT and 4DCBCT correspondence models. Radiotherapy and Oncology, 182, S1668–S1669.
  Büttgen, L.; Wimmert, L.; Madesta, F.; Werner, R. & Gauer, T.
  
• 1371: Impact of breathing signal-guided dose modulation on 4DCT image reconstruction. Radiotherapy and Oncology, 194, S3833-S3835.
  Wimmert, Lukas; Schwarz, Annette; Gauer, Tobias; Hofmann, Chrisitan; Dickmann, Jannis & Werner, Rene
  
• 668: Daily 4DCBT-based dose accumulation in adaptive 4D lung SBRT. Radiotherapy and Oncology, 194, S3988-S3990.
  Büttgen, Laura; Sentker, Thilo; Werner, Rene & Gauer, Tobias
  
• Benchmarking machine learning‐based real‐time respiratory signal predictors in 4D SBRT. Medical Physics, 51(5), 3173-3183.
  Wimmert, Lukas; Nielsen, Maximilian; Madesta, Frederic; Gauer, Tobias; Hofmann, Christian & Werner, Rene
  
• Impact of breathing signal‐guided dose modulation on step‐and‐shoot 4D CT image reconstruction. Medical Physics, 51(10), 7119-7126.
  Wimmert, Lukas; Schwarz, Annette; Gauer, Tobias; Hofmann, Christian; Dickmann, Jannis; Sentker, Thilo & Werner, Rene
  
• Monte Carlo-based simulation of virtual 3 and 4-dimensional cone-beam computed tomography from computed tomography images: An end-to-end framework and a deep learning-based speedup strategy. Physics and Imaging in Radiation Oncology, 32, 100644.
  Madesta, Frederic; Sentker, Thilo; Rohling, Clemens; Gauer, Tobias; Schmitz, Rüdiger & Werner, René
  
• Oriented histogram-based vector field embedding for characterizing 4D CT data sets in radiotherapy
  Madesta F., Wimmert L., Gauer T., Werner R. & Sentker T.
  
• Stability analysis of patient‐specific 4DCT‐ and 4DCBCT‐based correspondence models. Medical Physics, 51(9), 5890-5900.
  Büttgen, Laura Esther; Werner, René & Gauer, Tobias