Deep machine learning-assisted multiphoton microscopy to reduce light exposure and expedite imaging

Two-photon excitation fluorescence (2PEF) allows imaging of tissue up to about one millimeter in thickness. Typically, reducing fluorescence excitation exposure reduces the quality of the image. However, using deep learning super resolution techniques, these low-resolution images can be converted to high-resolution images. This work explores improving human tissue imaging by applying deep learning to maximize image quality while reducing fluorescence excitation exposure. We analyze two methods: a method based on U-Net, and a patch-based regression method. Both methods are evaluated on a skin dataset and an eye dataset. The eye dataset includes 1200 paired high power and low power images of retinal organoids. The skin dataset contains multiple frames of each sample of human skin. High-resolution images were formed by averaging 70 frames for each sample and low-resolution images were formed by averaging the first 7 and 15 frames for each sample. The skin dataset includes 550 images for each of the resolution levels. We track two measures of performance for the two methods: mean squared error (MSE) and structural similarity index measure (SSIM). For the eye dataset, the patches method achieves an average MSE of 27,611 compared to 146,855 for the U-Net method, and an average SSIM of 0.636 compared to 0.607 for the U-Net method. For the skin dataset, the patches method achieves an average MSE of 3.768 compared to 4.032 for the U-Net method, and an average SSIM of 0.824 compared to 0.783 for the U-Net method. Despite better performance on image quality, the patches method is worse than the U-Net method when comparing the speed of prediction, taking 303 seconds to predict one image compared to less than one second for the U-Net method.

[1]  Min-Ying Su,et al.  A multi-resolution approach for spinal metastasis detection using deep Siamese neural networks , 2017, Comput. Biol. Medicine.

[2]  Yoshiaki Kiuchi,et al.  Validation of a Deep Learning Model to Screen for Glaucoma Using Images from Different Fundus Cameras and Data Augmentation. , 2019, Ophthalmology. Glaucoma.

[3]  Jennifer J. Hunter,et al.  Safety assessment in macaques of light exposures for functional two-photon ophthalmoscopy in humans. , 2016, Biomedical optics express.

[4]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[5]  P. Baldi,et al.  Deep Learning Localizes and Identifies Polyps in Real Time With 96% Accuracy in Screening Colonoscopy. , 2018, Gastroenterology.

[6]  Pierre Baldi,et al.  Deep Learning in Biomedical Data Science , 2018, Annual Review of Biomedical Data Science.

[7]  Subhashini Venugopalan,et al.  Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs. , 2016, JAMA.

[8]  Eero P. Simoncelli,et al.  Image quality assessment: from error visibility to structural similarity , 2004, IEEE Transactions on Image Processing.

[9]  Pierre Baldi,et al.  Neural Networks for Fingerprint Recognition , 1993, Neural Computation.

[10]  Jürgen Schmidhuber,et al.  Deep learning in neural networks: An overview , 2014, Neural Networks.

[11]  Bruce J. Tromberg,et al.  In vivo multiphoton‐microscopy of picosecond‐laser‐induced optical breakdown in human skin , 2017, Lasers in surgery and medicine.

[12]  Yibo Zhang,et al.  Deep learning enhanced mobile-phone microscopy , 2017, ACS Photonics.

[13]  Scott E. Fraser,et al.  Structural and Functional Characterization of Human Stem-Cell-Derived Retinal Organoids by Live Imaging , 2017, Investigative ophthalmology & visual science.

[14]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[15]  Sabee Molloi,et al.  Detecting Cardiovascular Disease from Mammograms With Deep Learning , 2017, IEEE Transactions on Medical Imaging.

[16]  R. Dobarzić,et al.  [Fluorescence microscopy]. , 1975, Plucne bolesti i tuberkuloza.

[17]  A. Ozcan,et al.  Deep learning enables cross-modality super-resolution in fluorescence microscopy , 2018, Nature Methods.

[18]  Peter T C So,et al.  Antecedents of two‐photon excitation laser scanning microscopy , 2004, Microscopy research and technique.

[19]  Gregor Urban,et al.  Combining Deep Learning With Optical Coherence Tomography Imaging to Determine Scalp Hair and Follicle Counts , 2020, Lasers in surgery and medicine.

[20]  Bryan W. Jones,et al.  Photoreceptor Outer Segment-like Structures in Long-Term 3D Retinas from Human Pluripotent Stem Cells , 2017, Scientific Reports.

[21]  Maciej Wojtkowski,et al.  Two-photon imaging of the mammalian retina with ultrafast pulsing laser. , 2018, JCI insight.

[22]  M. Göppert-Mayer,et al.  Elementary processes with two quantum transitions , 2009 .

[23]  Rafael Yuste,et al.  Fluorescence microscopy today , 2005, Nature Methods.

[24]  Luca Maria Gambardella,et al.  Deep Neural Networks Segment Neuronal Membranes in Electron Microscopy Images , 2012, NIPS.

[25]  Pierre Baldi,et al.  Deep Learning for Drug Discovery and Cancer Research: Automated Analysis of Vascularization Images , 2019, IEEE/ACM Transactions on Computational Biology and Bioinformatics.

[26]  Ahmed Hosny,et al.  Data Analysis Strategies in Medical Imaging , 2018, Clinical Cancer Research.

[27]  Loic A. Royer,et al.  Content-aware image restoration: pushing the limits of fluorescence microscopy , 2018, Nature Methods.

[28]  P. Baldi,et al.  Deep-Learning Convolutional Neural Networks Accurately Classify Genetic Mutations in Gliomas , 2018, American Journal of Neuroradiology.

[29]  Elias T. Zambidis,et al.  Generation of three dimensional retinal tissue with functional photoreceptors from human iPSCs , 2014, Nature Communications.

[30]  S. K. Cheong,et al.  Selective S Cone Damage and Retinal Remodeling Following Intense Ultrashort Pulse Laser Exposures in the Near-Infrared , 2018, Investigative ophthalmology & visual science.