Yiting Wang

Review of spectral imaging technology in biomedical engineering: achievements and challenges

Abstract. Spectral imaging is a technology that integrates conventional imaging and spectroscopy to get both spatial and spectral information from an object. Although this technology was originally developed for remote sensing, it has been extended to the biomedical engineering field as a powerful analytical tool for biological and biomedical research. This review introduces the basics of spectral imaging, imaging methods, current equipment, and recent advances in biomedical applications. The performance and analytical capabilities of spectral imaging systems for biological and biomedical imaging are discussed. In particular, the current achievements and limitations of this technology in biomedical engineering are presented. The benefits and development trends of biomedical spectral imaging are highlighted to provide the reader with an insight into the current technological advances and its potential for biomedical research.

AOTF based molecular hyperspectral imaging system and its applications on nerve morphometry

The neuroanatomical morphology of nerve fibers is an important description for understanding the pathological aspects of nerves. Different from the traditional automatic nerve morphometry methods, a molecular hyperspectral imaging system based on an acousto-optic tunable filter (AOTF) was developed and used to identify unstained nerve histological sections. The hardware, software, and system performance of the imaging system are presented and discussed. The gray correction coefficient was used to calibrate the systems spectral response and to remove the effects of noises and artifacts. A spatial-spectral kernel-based approach through the support vector machine formulation was proposed to identify nerve fibers. This algorithm can jointly use both the spatial and spectral information of molecular hyperspectral images for segmentation. Then, the morphological parameters such as fiber diameter, axon diameter, myelin sheath thickness, fiber area, and g-ratio were calculated and evaluated. Experimental results show that the hyperspectral-based method has the potential to recognize and measure the nerve fiber more accurately than traditional methods.

Tongue fissure extraction and classification using hyperspectral imaging technology

Tongue fissures, an important feature on the tongue surface, may be pathologically related to some diseases. Most existing tongue fissure extraction methods use tongue images captured by traditional charge coupled device cameras. However, these conventional methods cannot be used for an accurate analysis of the tongue surface due to limited information from the images. To solve this, a hyperspectral tongue imager is used to capture tongue images instead of a digital camera. New algorithms for automatic tongue fissure extraction and classification, based on hyperspectral images, are presented. Both spectral and spatial information of the tongue surface is used to segment the tongue body and extract tongue fissures. Then a classification algorithm based on a hidden Markov model is used to classify tongue fissures into 12 typical categories. Results of the experiment show that the new method has good performance in terms of the classification rates of correctness.

Automatic identification and quantitative morphometry of unstained spinal nerve using molecular hyperspectral imaging technology

Quantitative observation of nerve fiber sections is often complemented by morphological analysis in both research and clinical condition. However, existing manual or semi-automated methods are tedious and labour intensive, fully automated morphometry methods are complicated as the information of color or gray images captured by traditional microscopy is limited. Moreover, most of the methods are time-consuming as the nerve sections need to be stained with some reagents before observation. To overcome these shortcomings, a molecular hyperspectral imaging system is developed and used to observe the spinal nerve sections. The molecular hyperspectral images contain both the structural and biochemical information of spinal nerve sections which is very useful for automatic identification and quantitative morphological analysis of nerve fibers. This characteristic makes it possible for researchers to observe the unstained spinal nerve and live cells in their native environment. To evaluate the performance of the new method, the molecular hyperspectral images were captured and the improved spectral angle mapper algorithm was proposed and used to segment the myelin contours. Then the morphological parameters such as myelin thickness and myelin area were calculated and evaluated. With these morphological parameters, the three dimension surface view images were drawn to help the investigators observe spinal nerve at different angles. The experiment results show that the hyperspectral based method has the potential to identify the spinal nerve more accurate than the traditional method as the new method contains both the spectral and spatial information of nerve sections.

Sublingual vein extraction algorithm based on hyperspectral tongue imaging technology

Among the parts of the human tongue surface, the sublingual vein is one of the most important ones which may have pathological relationship with some diseases. To analyze this information quantitatively, one primitive work is to extract sublingual veins accurately from tongue body. In this paper, a hyperspectral tongue imaging system instead of a digital camera is used to capture sublingual images. A hidden Markov model approach is presented to extract the sublingual veins from the hyperspectral sublingual images. This approach characterizes the spectral correlation and the band-to-band variability using a hidden Markov process, where the model parameters are estimated by the spectra of the pixel vectors forming the observation sequences. The proposed algorithm, the pixel-based sublingual vein segmentation algorithm, and the spectral angle mapper algorithm are tested on a total of 150 scenes of hyperspectral sublingual veins images to evaluate the performance of the new method. The experimental results demonstrate that the proposed algorithm can extract the sublingual veins more accurately than the traditional algorithms and can perform well even in a noisy environment.

AOTF based hyperspectral tongue imaging system and its applications in computer-aided tongue disease diagnosis

An AOTF-based hyperpsectral tongue imaging system which can capture both spectral and spatial information of human tongue is developed and used in computerized tongue diagnosis. This new type of tongue imaging system has the advantage of having no moving parts and can be scanned at very high rates. This makes the structure of the new system more compact and more suitable for tongue images capture. A series of hyperspectral images of tongue surface and sublingual veins are captured by using the new system. The automatic tongue segmentation and tongue features extraction based on hyperspectral images are also discussed. Preliminary experiment results show a great deal of potential of this system for computerized tongue diagnosis.

Leukocyte cells identification and quantitative morphometry based on molecular hyperspectral imaging technology

Leukocyte cells identification is one of the most frequently performed blood tests and plays an important role in the diagnosis of diseases. The quantitative observation of leukocyte cells is often complemented by morphological analysis in both research and clinical condition. Different from the traditional leukocyte cells morphometry methods, a molecular hyperspectral imaging system based on acousto-optic tunable filter (AOTF) was developed and used to observe the blood smears. A combined spatial and spectral algorithm is proposed to identify the cytoplasm and the nucleus of leukocyte cells by integrating the fuzzy C-means (FCM) with the spatial K-means algorithm. Then the morphological parameters such as the cytoplasm area, the nuclear area, the perimeter, the nuclear ratio, the form factor, and the solidity were calculated and evaluated. Experimental results show that the proposed algorithm has better performance than the spectral based algorithm as the new algorithm can jointly use the spatial and spectral information of leukocyte cells.

Pathological leucocyte segmentation algorithm based on hyperspectral imaging technique

White blood cells (WBC) are comparatively significant components in the human blood system, and they have a pathological relationship with some blood-related diseases. To analyze the disease information accurately, the most essential work is to segment WBCs. We propose a new method for pathological WBC segmentation based on a hyperspectral imaging system. This imaging system is used to capture WBC images, which is characterized by acquiring 1-D spectral information and 2-D spatial information for each pixel. A spectral information divergence algorithm is presented to segment pathological WBCs into four parts. In order to evaluate the performance of the new approach, K -means and spectral angle mapper-based segmental methods are tested in contrast on six groups of blood smears. Experimental results show that the presented method can segment pathological WBCs more accurately, regardless of their irregular shapes, sizes, and gray-values.

Molecular Spectral Imaging System for Quantitative Immunohistochemical Analysis of Early Diabetic Retinopathy

A molecular spectral imaging system has been developed based on microscopy and spectral imaging technology. The system is capable of acquiring molecular spectral images from 400 nm to 800 nm with 2 nm wavelength increments. The basic principles, instrumental systems, and system calibration method as well as its applications for the calculation of the stain-uptake by tissues are introduced. As a case study, the system is used for determining the pathogenesis of diabetic retinopathy and evaluating the therapeutic effects of erythropoietin. Some molecular spectral images of retinal sections of normal, diabetic, and treated rats were collected and analyzed. The typical transmittance curves of positive spots stained for albumin and advanced glycation end products are retrieved from molecular spectral data with the spectral response calibration algorithm. To explore and evaluate the protective effect of erythropoietin (EPO) on retinal albumin leakage of streptozotocin-induced diabetic rats, an algorithm based on Beer–Lamberts law is presented. The algorithm can assess the uptake by histologic retinal sections of stains used in quantitative pathology to label albumin leakage and advanced glycation end products formation. Experimental results show that the system is helpful for the ophthalmologist to reveal the pathogenesis of diabetic retinopathy and explore the protective effect of erythropoietin on retinal cells of diabetic rats. It also highlights the potential of molecular spectral imaging technology to provide more effective and reliable diagnostic criteria in pathology.

Histological skin morphology enhancement base on molecular hyperspectral imaging technology

Most traditional skin histological analysis methods are based on the light microscopy images, which can only provide limited information and low contrast results for pathology evaluation. Molecular hyperspectral imaging technology can provide both spatial and spectral information of skin sections, which is a new method for histological skin analysis.