DL4FLI

Deep Learning for Fluorescence Lifetime Imaging (FLI)

---

This GitHub contains relevant script, data and instructions for:

1. FLI data simulation workflow (MATLAB)

• FLIM '/TPSFsimulation_FLIM'

• Basic MFLI '/TPSFsimulation_basic'

• Monte-Carlo (via MCX) modeling of fluorescence decays through turbid media. THIS WORK IS ONGOING.

2. FLI-Net (neural network) training (python, [Tensorflow & Keras]).

• 3D-CNN for MFLI and FLIM analysis '/FLINET_ex'

3. SPCImage export and analysis instructions (general) along with example data. '/SPCImage_ExportAndAnalyze'

Authors: Jason T. Smith, Dr. Ruoyang Yao, Nathan Un, Dr. Pingkun Yan

Research Group: Functional & Molecular Optical Imaging Laboratory (RPI)

---

Relevant Data files:

MFLI Data:

1. Time-series MFLI for pharmacokinetic monitoring in vivo control & FRET-induced

2. NIR-FRET well-plate (AF700 & AF750)

3. In vivo matrigel ROIs for NIR FRET quantification (AF700 & AF750)

4. MFLI acquisitions of well-plates containing serial dilutions of ATTO 740 & HITCI (methodology detailed here (REFERENCE))

5. Serial dilution of AF750 imaged at 25mW & 75mW laser power for performance assessment at low photon-counts

Visible FLIM Data

1. Visible FRET-FLIM (AF488 & AF555, T47D breast cancer cells)

2. Visible Metabolic NAD(P)H FLIM pre & post exposure to sodium cyanide. The breast cancer cell lines used include the following:

• MCF10a
• AU565
• T47D
• MDA-MB 231

NIR FLIM Data

1. NIR FRET-FLIM (AF700 & AF750, T47D breast cancer cells)

In Silico Data

1. Data simulated across three photon count thresholds (25-100, 100-250 & 250-500)

---

Related Publications/Presentations:

PNAS Manuscript & Supplementary Information

1. Smith JT, Yao R, Sinsuebphon N, Rudkouskaya A, Un N, Mazurkiewicz J, Barroso M, Yan P, Intes X. Fast fit-free analysis of fluorescence lifetime imaging via deep learning. Proceedings of the National Academy of Sciences. 2019 Nov 12.

Original preprint

2. Smith JT, Yao R, Sinsuebphon N, Rudkouskaya A, Mazurkiewicz J, Barroso M, Yan P, Intes X. Ultra-fast fit-free analysis of complex fluorescence lifetime imaging via deep learning. bioRxiv. 2019 Jan 1:523928.

OSA Proceeding

3. Smith JT, Un N, Yao R, Sinsuebphon N, Rudkouskaya A, Mazurkiewicz J, Barroso M, Yan P, Intes X. Fluorescent Lifetime Imaging improved via Deep Learning. InNovel Techniques in Microscopy 2019 Apr 14 (pp. NM3C-4). Optical Society of America.

SPIE Presentation

4. Smith JT, Yao R, Chen SJ, Sinsuebphon N, Rudkouskaya A, Barroso M, Yan P, Intes X. Deep learning for quantitative bi-exponential fluorescence lifetime imaging (Conference Presentation). InMultimodal Biomedical Imaging XIV 2019 Mar 4 (Vol. 10871, p. 108710J). International Society for Optics and Photonics.

---