FLIMAnalyzer

Probing cellular metabolic states with fluorescence lifetime imaging microscopy

FLIMAnalyzer is a recently published Python package, developed in collaboration with Dr. Ammasi Periasamy’s group (Keck Center for Cellular Imaging) to support fluorescence lifetime imaging microscopy (FLIM) analysis. It applies ML/DL techniques to enhance the sensitivity of detecting changes in cellular redox states. Designed as an extensible analysis framework, the package includes plugins for importing multi-parameter FLIM data from tabular CSV files, filtering and organizing the data in spreadsheet format, performing statistical analyses, and visualizing data. Deep learning modules with PyTorch-based autoencoders enable feature extraction and data augmentation, including built-in support for hyperparameter tuning. The modular architecture allows for the addition of new plugins and the composition of custom, multi-stage analysis workflows by chaining plugin-defined tasks into a directed acyclic graph. As with MitosisAnalyzer, this package leverages Prefect for task orchestration, with Dask available for distributed execution at scale. Although originally developed for FLIM data analysis, the framework is adaptable to other multi-feature datasets structured in tabular format.

Collaborators: Ammasi Periasamy, Horst Wallrabe, Shagufta Rehman Alam, Brian Mbogo, Jianxin Zhang, Thomas Cassidy

Publications

Label-free optical biomarker for prostate cancer diagnosis Shagufta Rehman Alam, Karsten Siller, Robert Dreicer, Ammasi Periasamy Molecular Medicine (2026)
Measuring metabolic changes in cancer cells using two‐photon fluorescence lifetime imaging microscopy and machine‐learning analysis Jiaxin Zhang, Horst Wallrabe, Karsten Siller, Brian Mbogo, Thomas Cassidy, Shagufta Rehman Alam, Ammasi Periasamy Journal of Biophotonics, 18(1), e202400426 (2025)

Code: https://github.com/uvaKCCI/flimanalyzer

Funding: Internal seed grant