Raman spectromics method for fast and label-free genotype screening

It is now understood that genes and their various mutations are associated with the onset and progression of diseases. However, routine genetic testing techniques are limited by their high cost, time consumption, susceptibility to contamination, complex operation, and data analysis difficulties, rendering them unsuitable for genotype screening in many cases. Therefore, there is an urgent need to develop a rapid, sensitive, user-friendly, and cost-effective method for genotype screening and analysis. In this study, we propose and investigate a Raman spectroscopic method for achieving fast and label-free genotype screening. The method was validated using spontaneous Raman measurements of wild-type Cryptococcus neoformans and its six mutants. An accurate identification of different genotypes was achieved by employing a one-dimensional convolutional neural network (1D-CNN), and significant correlations between metabolic changes and genotypic variations were revealed. Genotype-specific regions of interest were also localized and visualized using a gradient-weighted class activation mapping (Grad-CAM)-based spectral interpretable analysis method. Furthermore, the contribution of each metabolite to the final genotypic decision-making was quantified. The proposed Raman spectroscopic method demonstrated huge potential for fast and label-free genotype screening and analysis of conditioned pathogens.

Early assessment of chemotherapeutic response in hepatocellular carcinoma based on serum surface-enhanced Raman spectroscopy

In clinical practice, the transcatheter arterial chemoembolization (TACE) has been widely accepted as the first option for non-surgical hepatocellular carcinoma (HCC) treatment. However, patients with HCC often suffer from poor response to TACE therapy. This can be prevented if the chemotherapeutic response can be early and accurately assessed, which is essential to guide timely and rational management. In this study, the serum SERS technique was for the first time investigated as a potential prognostic tool for early assessment of HCC chemotherapeutic response. According to the SERS spectral analysis results, it is newly found that not only the absolute circulating nucleic acids and collagen levels in pre-therapeutic serum but also the changes in circulating nucleic acids and amino acids between pre-therapeutic and post-therapeutic serum are expected to be potential serum markers for HCC prognosis. By further applying chemometrics methods to establish prognostic models, excellent prognostic accuracies were achieved within only 3 days after TACE therapy. Thus, the proposed method is expected to provide guidance on timely and rational management of HCC to improve its survival rate.

Prediction of the postoperative prognosis in patients with non-muscle-invasive bladder cancer based on preoperative serum surface-enhanced Raman spectroscopy

Non-muscle-invasive bladder cancer (NMIBC) is a common urinary tumor and has a high recurrence rate due to improper or inadequate conservative treatment. The early and accurate prediction of its recurrence can be helpful to implement timely and rational treatment. In this study, we explored a preoperative serum surface-enhanced Raman spectroscopy based prognostic protocol to predict the postoperative prognosis for NMIBC patients at the time even before treatment. The biochemical analysis results suggested that biomolecules related to DNA/RNA, protein substances, trehalose and collagen are expected to be potential prognostic markers, which further compared with several routine clinically used immunohistochemistry expressions with prognostic values. In addition, high prognostic accuracies of 87.01% and 89.47% were achieved by using the proposed prognostic models to predict the future postoperative recurrence and recurrent type, respectively. Therefore, we believe that the proposed method has great potential in the early and accurate prediction of postoperative prognosis in patients with NMIBC, which is with important clinical significance to guide the treatment and further improve the recurrence rate and survival time.

Tortuosity of branch retinal artery is more associated with the genesis and progress of diabetic retinopathy

OBJECTIVE

The purpose of this study is to investigate the potential of using the tortuosity of branch retinal artery as a more promising indicator for early detection and accurate assessment of diabetic retinopathy (DR).

DESIGN AND METHOD

The diagnoses, consisting of whether DR or not as well as DR severity, were given by ophthalmologists upon the assessment of those fundus images from 495 diabetic patients. Meanwhile, benefiting from those good contrast and high optical resolution fundus images taken by confocal scanning laser ophthalmoscope, the branch arteries, branch veins, main arteries and main veins in retina can be segmented independently, and the tortuosity values of them were further extracted to investigate their potential correlations with DR genesis and progress based on one-way ANOVA test.

RESULTS

For both two comparisons, i.e., between non-DR group and DR group as well as among groups with different DR severity levels, larger tortuosity increments were always observed in retinal arteries and the increments in branch retinal vessels were even larger. Furthermore, it was newly found that branch arterial tortuosity was significantly associated with both DR genesis (p=0.030) and DR progress (p<0.001).

CONCLUSION

Based on this cohort study of 495 diabetic patients without DR and with different DR severity, the branch arterial tortuosity has been found to be more closely associated with DR genesis as well as DR progress. Therefore, the branch arterial tortuosity is expected to be a more direct and specific indicator for early detection of DR as well as accurate assessment of DR severity, which can further guide timely and rational management of DR to prevent from visual impairment or even blindness resulting from DR.

Identification and assessment of pulmonary Cryptococcus neoformans infection by blood serum surface-enhanced Raman spectroscopy

Cryptococcus neoformans (C. neoformans) is a causative agent for acute pulmonary infection, which can further develop to lethal meningoencephalitis if untreated. The meningoencephalitis infection can be prevented, if timely treatment on pulmonary cryptococcal infection can be implemented based on its early diagnosis and accurate assessment. In this study, blood serum surface-enhanced Raman spectroscopy (SERS) method was investigated on identification and assessment of pulmonary C. neoformans infection. The serum SERS measurements were collected from the mice infected with C. neoformans and the healthy mice, in which the infected mice were further divided into four subgroups according to the duration of infection. Based on those SRES measurements, biochemical differences were analyzed among those different groups to investigate the potential biomarkers for identifying and assessing the pulmonary C. neoformans infection. Furthermore, partial least square (PLS) analysis followed by linear discriminant analysis (LDA) model was employed to identify pulmonary cryptococcal infection and to assess the degrees of infection with the accuracies of 96.7% and 85.3%, respectively. Therefore, our study has demonstrated the great clinical potential of using serum SERS technique for an accurate identification and assessment of pulmonary cryptococcal infection.

Programmable hyperspectral microscopy for high-contrast biomedical imaging in a snapshot

SIGNIFICANCE

Hyperspectral microscopy has been intensively explored in biomedical applications. However, due to its huge three-dimensional hyperspectral data cube, it typically suffers from slow data acquisition, mass data transmission and storage, and computationally expensive postprocessing.

AIM

To overcome the above limitations, a programmable hyperspectral microscopy technique was developed, which can perform hardware-based hyperspectral data postprocessing by the physical process of optical imaging in a snapshot.

APPROACH

A programmable hyperspectral microscopy system was developed to collect coded microscopic images from samples under multiplexed illumination. Principal component analysis followed by linear discriminant analysis scheme was coded into multiplexed illumination and realized by the physical process of optical imaging. The contrast enhancement was evaluated on two representative types of microscopic samples, i.e., tissue section and cell samples.

RESULTS

Compared to the microscopic images collected under white light illumination, the contrasts of coded microscopic images were significantly improved by 41% and 59% for tissue section and cell samples, respectively.

CONCLUSIONS

The proposed method can perform hyperspectral data acquisition and postprocessing simultaneously by its physical process, while preserving the most important spectral information to maximize the difference between the target and background, thus opening a new avenue for high-contrast microscopic imaging in a snapshot.

Predicting Unnecessary Nodule Biopsies from a Small, Unbalanced, and Pathologically Proven Dataset by Transfer Learning

This study explores an automatic diagnosis method to predict unnecessary nodule biopsy from a small, unbalanced, and pathologically proven database. The automatic diagnosis method is based on a convolutional neural network (CNN) model. Because of the small and unbalanced samples, the presented method aims to improve the transfer learning capability via the VGG16 architecture and optimize the related transfer learning parameters. For comparison purpose, a traditional machine learning method is implemented, which extracts the texture features and classifies the features by support vector machine (SVM). The database includes 68 biopsied nodules, 16 are pathologically proven benign and the remaining 52 are malignant. To consider the volumetric data by the CNN model, each image slice from each nodule volume is selected randomly until all image slices of each nodule are utilized. The leave-one-out and 10-folder cross validations are applied to train and test the randomly selected 68 image slices (one image slice from one nodule) in each experiment, respectively. The averages over all the experimental outcomes are the final results. The experiments revealed that the features from both the medical and the natural images share the similarity of focusing on simpler and less-abstract objects, leading to the conclusion that not the more the transfer convolutional layers, the better the classification results. Transfer learning from other larger datasets can supply additional information to small and unbalanced datasets to improve the classification performance. The presented method has shown the potential to adapt CNN architecture to improve the prediction of unnecessary nodule biopsy from small, unbalanced, and pathologically proven volumetric dataset.