Fast Analysis of Time-Domain Fluorescence Lifetime Imaging via Extreme Learning Machine

Abstract

We present a fast and accurate analytical method for fluorescence lifetime imaging microscopy (FLIM), using the extreme learning machine (ELM). We used extensive metrics to evaluate ELM and existing algorithms. First, we compared these algorithms using synthetic datasets. The results indicate that ELM can obtain higher fidelity, even in low-photon conditions. Afterwards, we used ELM to retrieve lifetime components from human prostate cancer cells loaded with gold nanosensors, showing that ELM also outperforms the iterative fitting and non-fitting algorithms. By comparing ELM with a computational efficient neural network, ELM achieves comparable accuracy with less training and inference time. As there is no back-propagation process for ELM during the training phase, the training speed is much higher than existing neural network approaches. The proposed strategy is promising for edge computing with online training.

Simple and Robust Deep Learning Approach for Fast Fluorescence Lifetime Imaging

“…Another MLP [30] processes phasor coordinates and generates lifetime parameters. Because the inputs are coordinates rather than histograms, the MLP 3-D CNN [31] 1-D CNN [25] 1-D CNN [28] MLP [27] Phasor-MLP [30] MLP [29] GAN [33] ELM [36] FLAN FLAN+LS The model size was not given in the paper; the size was estimated from the network's structure. exhibits fast speed.…”

“…To enhance the spatial resolution of FLIM images, Dong et al [35] introduced a cascade CNN architecture to infer fluorescence lifetime and improve the spatial resolution afterwards. An Extreme Learning Machine (ELM) [36] was presented to achieve fast and accurate lifetimes reconstruction with back-propagation-free, online training.…”

“…Despite the simplicity, reconstructed lifetime parameters sometimes had distinct dispersion than 3-D CNN [31]. ELM [36] has the simplest structure containing only one hidden layer.…”

“…A general structure of a deep-learning-based predictive model is shown in Figure 1 a, including the model training from training data in phase 1 and model testing from new data in phase 2. Various neural networks were proposed for multiexponential analysis, including multilayer perceptions (MLPs) [ 26 ], one- and high-dimensional convolutional neural networks (CNNs) [ 27 ], online training extreme learning machine [ 28 ], and generative adversarial networks (GANs) [ 29 ]. MLP is simple and can be easily implemented on different platforms, such as CPU, GPU, or other embedded systems.…”

Deep‐learning FRET visualization in flow cytometry: At the cross road of the signaling and FRET pathways

“…Such methods start to become quite widespread today. They are called collectively as "artificial intelligence" techniques, referring mainly to "deep-learning," (or "soft computing," "machine-vision"), and "reduction of dimensionality" [5][6][7][8][9][10][11].…”