Current Research

AI for Clinic

Multiple Brain Metastases Management Platform

Stereotactic radiosurgery (SRS) has become a primary treatment for multiple brain metastases (BM) but may require the distribution of BMs over several sessions to make delivery time and radiation toxicity manageable. Contrasting to equal fraction dose in conventional fractionation, distributed SRS delivers the full dose to a subset of BMs in each session while avoiding adjacent BMs in the same session to reduce toxicity from overlapping radiation. However, current clinical treatment planning for distributed SRS relies on manual BM assignment, which can be tedious and error-prone. Novel integration of AI and machine learning in segmentation and registration/tracking are potential clinical solutions for improving multiple BM management.

Our lab focuses on building an AimBMs platform with novel AI technologies:

  1. to accurately and efficiently auto-segmenting and distributing multiple BMs
  2. for developing multimodal AI model to predict SRS outcomes for better patient care.

Related Publications

1. Yang, Z., Liu, H., Liu, Y., Stojadinovic, S., Timmerman, R., Nedzi, L., Dan, T., Wardak, Z., Lu, W. and Gu, X. (2020), A web-based brain metastases segmentation and labeling platform for stereotactic radiosurgery. Med. Phys., 47: 3263-3276. https://doi.org/10.1002/mp.14201

2. Chen, M., Wardak, Z., Stojadinovic, S., Gu, X. and Lu, W. (2021), A general algorithm for distributed treatments of multiple brain metastases. Med. Phys., 48: 1832-1838. https://doi.org/10.1002/mp.14722

3. Yang, Z., Chen, M., Kazemimoghadam, M, Ma, L., Stajoadinovic, S., Timmerman, R., Dan, T., Wardak, Z., Lu, W. and Gu, X. (2022), Deep-learning and radiomics ensemble classifier for false positive reduction in brain metastases segmentation. Phys. Med. Biol., 67: 025004. https://doi.org/10.1088/1361-6560/ac4667

AI for Preclinic

AI-Empowered Crypt Scoring & Iso-Effective Dose Estimation Platform

The crypts of the small intestine are pivotal for studying radiation-induced damage and tissue recovery due to their high proliferation rate. Quantifying and analyzing crypt microcolony in histology image will eclucidate tissue radiation-response across different radiation modality for radiotherapy protocol optimization.

Our lab focuses on building an AI-Crypt platform using novel AI technologies to:

  1. accurately and efficiently identify crypts
  2. analyze crypt microcolony for iso-effective radiation dose estimation

 

Related Publications

1.  Fu J., et al. Exploring Deep Learning for Estimating the Isoeffective Dose of FLASH Irradiation From Mouse Intestinal Histological Images. International Journal of Radiation Oncology*Biology*Physics. 2024;119(3):1001-1010, DOI 10.1016/j.ijrobp.2023.12.032