Monitoring Amyloidogenesis with a 3D Deep-Learning-Guided Biolaser Imaging Array.

Amyloidogenesis is a critical hallmark for many neurodegenerative diseases and drug screening; however, identifying intermediate states of protein aggregates at an earlier stage remains challenging. Herein, we developed a peptide-encapsulated droplet microlaser to monitor the amyloidogenesis process and evaluate the efficacy of anti-amyloid drugs. The lasing wavelength changes accordingly with the amyloid peptide folding behaviors and nanostructure conformations in the droplet resonator. A 3D deep-learning strategy was developed to directly image minute spectral shifts through a far-field camera. By extracting 1D color information and 2D features from the laser images, the progression of the amyloidogenesis process could be monitored using arrays of laser images from microdroplets. The training set, validation set, and test set of the multimodal learning model achieved outstanding classification accuracies of over 95%. This study shows the great potential of deep-learning-empowered peptide microlaser yields for protein misfolding studies and paves the way for new possibilities for high-throughput imaging of cavity biosensing.

Fluorescence imaging and Raman spectroscopy applied for the accurate diagnosis of breast cancer with deep learning algorithms.

Deep learning is usually combined with a single detection technique in the field of disease diagnosis. This study focused on simultaneously combining deep learning with multiple detection technologies, fluorescence imaging and Raman spectroscopy, for breast cancer diagnosis. A number of fluorescence images and Raman spectra were collected from breast tissue sections of 14 patients. Pseudo-color enhancement algorithm and a convolutional neural network were applied to the fluorescence image processing, so that the discriminant accuracy of test sets, 88.61%, was obtained. Two different BP-neural networks were applied to the Raman spectra that mainly comprised collagen and lipid, so that the discriminant accuracy of 95.33% and 98.67% of test sets were gotten, respectively. Then the discriminant results of fluorescence images and Raman spectra were counted and arranged into a characteristic variable matrix to predict the breast tissue samples with partial least squares (PLS) algorithm. As a result, the predictions of all samples are correct, with minor error of predictive value. This study proves that deep learning algorithms can be applied into multiple diagnostic optics/spectroscopy techniques simultaneously to improve the accuracy in disease diagnosis.