Multi-Scale Attention Convolutional Network for Masson Stained Bile Duct Segmentation from Liver Pathology Images.

In clinical practice, the Ishak Score system would be adopted to perform the evaluation of the grading and staging of hepatitis according to whether portal areas have fibrous expansion, bridging with other portal areas, or bridging with central veins. Based on these staging criteria, it is necessary to identify portal areas and central veins when performing the Ishak Score staging. The bile ducts have variant types and are very difficult to be detected under a single magnification, hence pathologists must observe bile ducts at different magnifications to obtain sufficient information. This pathologic examinations in routine clinical practice, however, would result in the labor intensive and expensive examination process. Therefore, the automatic quantitative analysis for pathologic examinations has had an increased demand and attracted significant attention recently. A multi-scale inputs of attention convolutional network is proposed in this study to simulate pathologists' examination procedure for observing bile ducts under different magnifications in liver biopsy. The proposed multi-scale attention network integrates cell-level information and adjacent structural feature information for bile duct segmentation. In addition, the attention mechanism of proposed model enables the network to focus the segmentation task on the input of high magnification, reducing the influence from low magnification input, but still helps to provide wider field of surrounding information. In comparison with existing models, including FCN, U-Net, SegNet, DeepLabv3 and DeepLabv3-plus, the experimental results demonstrated that the proposed model improved the segmentation performance on Masson bile duct segmentation task with 72.5% IOU and 84.1% F1-score.

Continuous Wave Spectroscopy with Diffusion Theory for Quantification of Optical Properties: Comparison Between Multi-distance and Multi-wavelength Data Fitting Methods.

Typically, continuous wave spectroscopy (CWS) can be used to accurately quantify biological tissue optical properties (µ a and µ s ') by employing the diffuse reflectance information acquired at multiple source-detector separations (multi-distance). On the other hand, sample optical properties can also be obtained by fitting multi-wavelength light reflectance acquired at a single source detector separation to the diffusion theory equation. To date, multi-wavelength and multi-distance methods have not yet been rigorously compared for their accuracy in quantification of the sample optical properties. In this investigation, we compared the accuracy of the two above-mentioned quantifying methods in the optical properties recovery. The liquid phantoms had µ a between 0.004 and 0.011 mm(-1) and µ s ' between 0.55 and 1.07 mm(-1) whose optical properties mimic the human breast. Multi-distance data and multi-wavelength data were fitted to the same diffusion equation for consistency. The difference between benchmark µ a and µ s ' and the fitted results, ΔError (ΔE) was used to evaluate the accuracy of the two methods. The results showed that either method yielded ΔE within 15-30 % when values were within certain limits to standard values applicable to µ s ' and µ a for human adipose tissue. Both methods showed no significant differences in ΔE values. Our results suggest that both multi-distance and multi-wavelength methods can yield similar reasonable optical properties in biological tissue with a proper calibration.

Enhancing Organizing Pneumonia Diagnosis: A Novel Super-token Transformer Approach for Masson Body Segmentation.

BACKGROUND/AIM: In this study, we introduce an innovative deep-learning model architecture aimed at enhancing the accuracy of detecting and classifying organizing pneumonia (OP), a condition characterized by the presence of Masson bodies within the alveolar spaces due to lung injury. The variable morphology of Masson bodies and their resemblance to adjacent pulmonary structures pose significant diagnostic challenges, necessitating a model capable of discerning subtle textural and structural differences. Our model incorporates a novel architecture that integrates advancements in three key areas: Semantic segmentation, texture analysis, and structural feature recognition. MATERIALS AND METHODS: We employed a dataset of whole slide imaging from 20 patients, totaling 100 slides of OP, segmented into training, validation, and testing sets to reflect real-world application scenarios. Our approach utilizes a modified multi-head self-attention mechanism combined with ResUNet for semantic segmentation, enhanced by superpixel concepts. This method facilitates the generation of representative token features through iterative super-token blocks, creating high-resolution token maps that leverage local and high-level feature information for improved accuracy. RESULTS: Benefiting from token features and distribution for enhanced texture alignment with fewer false-positives, the super-token transformer (STT) model achieved a mean intersection over union (mIOU) of 72.42%, with a sensitivity of 47.81%, specificity of 99.83%, positive predictive value of 64.03%, and negative predictive value of 99.94%, highlighting superior efficacy in Masson body segmentation in complex cross-tissue analyses. CONCLUSION: Our team developed an iterative learning model based on the STT approach, emphasizing token features of super token, including texture and distribution, that enable enhanced alignment with the unique textures of Masson bodies to improve sensitivity and mIOU, The development of this STT model presents a significant advancement in the field of medical image analysis for OP that offers a promising avenue for improving diagnostic precision and patient outcomes in pulmonary pathology.

Design of HL7 FHIR Profiles for Pathology Reports Integrated with Pathology Images.

This paper describes the development of Health Level Seven Fast Healthcare Interoperability Resource (FHIR) profiles for pathology reports integrated with whole slide images and clinical data to create a pathology research database. A report template was designed to collect structured reports, enabling pathologists to select structured terms based on a checklist, allowing for the standardization of terms used to describe tumor features. We gathered and analyzed 190 non-small-cell lung cancer pathology reports in free text format, which were then structured by mapping the itemized vocabulary to FHIR observation resources, using international standard terminologies, such as the International Classification of Diseases, LOINC, and SNOMED CT. The resulting FHIR profiles were published as an implementation guide, which includes 25 profiles for essential data elements, value sets, and structured definitions for integrating clinical data and pathology images associated with the pathology report. These profiles enable the exchange of structured data between systems and facilitate the integration of pathology data into electronic health records, which can improve the quality of care for patients with cancer.

Lymphocyte-Infiltrated Periportal Region Detection With Structurally-Refined Deep Portal Segmentation and Heterogeneous Infiltration Features.

Goal: The early diagnosis and treatment of hepatitis is essential to reduce hepatitis-related liver function deterioration and mortality. One component of the widely-used Ishak grading system for the grading of periportal interface hepatitis is based on the percentage of portal borders infiltrated by lymphocytes. Thus, the accurate detection of lymphocyte-infiltrated periportal regions is critical in the diagnosis of hepatitis. However, the infiltrating lymphocytes usually result in the formation of ambiguous and highly-irregular portal boundaries, and thus identifying the infiltrated portal boundary regions precisely using automated methods is challenging. This study aims to develop a deep-learning-based automatic detection framework to assist diagnosis. Methods: The present study proposes a framework consisting of a Structurally-REfined Deep Portal Segmentation module and an Infiltrated Periportal Region Detection module based on heterogeneous infiltration features to accurately identify the infiltrated periportal regions in liver Whole Slide Images. Results: The proposed method achieves 0.725 in F1-score of lymphocyte-infiltrated periportal region detection. Moreover, the statistics of the ratio of the detected infiltrated portal boundary have high correlation to the Ishak grade (Spearman's correlations more than 0.87 with p-values less than 0.001) and medium correlation to the liver function index aspartate aminotransferase and alanine aminotransferase (Spearman's correlations more than 0.63 and 0.57 with p-values less than 0.001). Conclusions: The study shows the statistics of the ratio of infiltrated portal boundary have correlation to the Ishak grade and liver function index. The proposed framework provides pathologists with a useful and reliable tool for hepatitis diagnosis.

Breast cancer detection of large size to DCIS by hypoxia and angiogenesis using NIRS.

This investigation aimed to test all tumor-bearing patients who undergo biopsy to see if angiogenesis and hypoxia can detect cancer. We used continuous-wave near-infrared spectroscopy (NIRS) to measure blood hemoglobin concentration to obtain blood volume or total hemoglobin [Hbtot] and oxygen saturation for the angiogenesis and hypoxic biomarkers. The contralateral breast was used as a reference to derive the difference from breast tumor as a difference in total hemoglobin Δ[HBtot] and a difference in deoxygenation Δ([Hb]-[HbO2]). A total of 91 invasive cancers, 26 DCIS, 45 fibroblastomas, 96 benign tumors excluding cysts, and 67 normal breasts were examined from four hospitals. In larger-size tumors, there is significantly higher deoxygenation in invasive and ductal carcinoma in situ (DCIS) than in that of benign tumors, but no significant difference was seen in smaller tumors of ≤ 1 cm. With the two parameters of high total hemoglobin and hypoxia score, the sensitivity and specificity of cancer detection were 60.3 % and 85.3 %, respectively. In summary, smaller-size tumors are difficult to detect with NIRS, whereas DCIS can be detected by the same total hemoglobin and hypoxic score in our study.

ADMM-SRNet: Alternating Direction Method of Multipliers Based Sparse Representation Network for One-Class Classification.

One-class classification aims to learn one-class models from only in-class training samples. Because of lacking out-of-class samples during training, most conventional deep learning based methods suffer from the feature collapse problem. In contrast, contrastive learning based methods can learn features from only in-class samples but are hard to be end-to-end trained with one-class models. To address the aforementioned problems, we propose alternating direction method of multipliers based sparse representation network (ADMM-SRNet). ADMM-SRNet contains the heterogeneous contrastive feature (HCF) network and the sparse dictionary (SD) network. The HCF network learns in-class heterogeneous contrastive features by using contrastive learning with heterogeneous augmentations. Then, the SD network models the distributions of the in-class training samples by using dictionaries computed based on ADMM. By coupling the HCF network, SD network and the proposed loss functions, our method can effectively learn discriminative features and one-class models of the in-class training samples in an end-to-end trainable manner. Experimental results show that the proposed method outperforms state-of-the-art methods on CIFAR-10, CIFAR-100 and ImageNet-30 datasets under one-class classification settings. Code is available at https://github.com/nchucvml/ADMM-SRNet.

Pathologic liver tumor detection using feature aligned multi-scale convolutional network.

The detection of the most common type of liver tumor, that is, hepatocellular carcinoma (HCC), is one essential step to liver pathology image analysis. In liver tissue, common cell change phenomena such as apoptosis, necrosis, and steatosis are similar in tumor and benign tissue. Hence, the detection of HCC may fail when the patches covered only limited tissue region without enough neighboring cell structure information. To address this problem, a Feature Aligned Multi-Scale Convolutional Network (FA-MSCN) architecture is proposed in this paper for automatic liver tumor detection based on whole slide images (WSI). The proposed network integrates the features obtained at different magnification levels to improve the detection performance by referencing more neighboring information. The FA-MSCN consists of two parallel convolutional networks in which one would extract high-resolution features and the other would extract low-resolution features by atrous convolution. The low-resolution features then go through central cropping, upsampling, and concatenation with high-resolution features for final classification. The experimental results demonstrated that Multi-Scale Convolutional Network (MSCN) improves the detection performance compared to Single-Scale Convolutional Network (SSCN), and that the FA-MSCN is superior to both SSCN and MSCN, demonstrating on HCC detection.

Texture-Map-Based Branch-Collaborative Network for Oral Cancer Detection.

The paper proposes an innovative deep convolutional neural network (DCNN) combined with texture map for detecting cancerous regions and marking the ROI in a single model automatically. The proposed DCNN model contains two collaborative branches, namely an upper branch to perform oral cancer detection, and a lower branch to perform semantic segmentation and ROI marking. With the upper branch the network model extracts the cancerous regions, and the lower branch makes the cancerous regions more precision. To make the features in the cancerous more regular, the network model extracts the texture images from the input image. A sliding window is then applied to compute the standard deviation values of the texture image. Finally, the standard deviation values are used to construct a texture map, which is partitioned into multiple patches and used as the input data to the deep convolutional network model. The method proposed by this paper is called texture-map-based branch-collaborative network. In the experimental result, the average sensitivity and specificity of detection are up to 0.9687 and 0.7129, respectively based on wavelet transform. And the average sensitivity and specificity of detection are up to 0.9314 and 0.9475, respectively based on Gabor filter.

Two-channel autofluorescence analysis for oral cancer.

We created a two-channel autofluorescence test to detect oral cancer. The wavelengths 375 and 460 nm, with filters of 479 and 525 nm, were designed to excite and detect reduced-form nicotinamide adenine dinucleotide (NADH) and flavin adenine dinucleotide (FAD) autofluorescence. Patients with oral cancer or with precancerous lesions, and a control group with healthy oral mucosae, were enrolled. The lesion in the autofluorescent image was the region of interest. The average intensity and heterogeneity of the NADH and FAD were calculated. The redox ratio [(NADH)/(NADH + FAD)] was also computed. A quadratic discriminant analysis (QDA) was used to compute boundaries based on sensitivity and specificity. We analyzed 49 oral cancer lesions, 34 precancerous lesions, and 77 healthy oral mucosae. A boundary (sensitivity: 0.974 and specificity: 0.898) between the oral cancer lesions and healthy oral mucosae was validated. Oral cancer and precancerous lesions were also differentiated from healthy oral mucosae (sensitivity: 0.919 and specificity: 0.755). The two-channel autofluorescence detection device and analyses of the intensity and heterogeneity of NADH, and of FAD, and the redox ratio combined with a QDA classifier can differentiate oral cancer and precancerous lesions from healthy oral mucosae.

Novel quantitative analysis of autofluorescence images for oral cancer screening.

OBJECTIVES: VELscope® was developed to inspect oral mucosa autofluorescence. However, its accuracy is heavily dependent on the examining physician's experience. This study was aimed toward the development of a novel quantitative analysis of autofluorescence images for oral cancer screening. MATERIALS AND METHODS: Patients with either oral cancer or precancerous lesions and a control group with normal oral mucosa were enrolled in this study. White light images and VELscope® autofluorescence images of the lesions were taken with a digital camera. The lesion in the image was chosen as the region of interest (ROI). The average intensity and heterogeneity of the ROI were calculated. A quadratic discriminant analysis (QDA) was utilized to compute boundaries based on sensitivity and specificity. RESULTS: 47 oral cancer lesions, 54 precancerous lesions, and 39 normal oral mucosae controls were analyzed. A boundary of specificity of 0.923 and a sensitivity of 0.979 between the oral cancer lesions and normal oral mucosae were validated. The oral cancer and precancerous lesions could also be differentiated from normal oral mucosae with a specificity of 0.923 and a sensitivity of 0.970. CONCLUSION: The novel quantitative analysis of the intensity and heterogeneity of VELscope® autofluorescence images used in this study in combination with a QDA classifier can be used to differentiate oral cancer and precancerous lesions from normal oral mucosae.

Reconstruction and rendering of microcalcifications from two mammogram views by modified projective grid space (MPGS).

Mammograms taken by two views: cranio-caudal (CC) and medio-lateral oblique (MLO) views provide only 2D projections of the microcalcifications, which lack the depth information. Thus, envisioning the relative lesion location from mammograms is a challenge for radiologists. To assist radiologists in locating and rendering lesion tissues, a modified projective grid space (MPGS) scheme is proposed to reconstruct 3D microcalcifications. The MPGS scheme reconstructs 3D microcalcifications in a unique space defined by corresponding points and the epipoles retrieved from the fundamental matrix of the CC and MLO views. Since only corresponding points of images are required in the proposed MPGS scheme, we can avoid the difficulty associated with most reconstruction approaches that require prior complicated calibration of X-ray machine. Considering the deformation of the breast, a new method based on the concept of bundle adjustment is proposed to rectify the 3D locations of reconstructed microcalcifications by uncompressed breast model reconstructed from the real patient body using MPGS scheme with iterative closest point (ICP). Then, the reconstructed microcalcifications are augmented in the real patient body model to show their relative positions.

A new method for adjusting neural response jitter in the STRF obtained by spike-trigger averaging.

Sensitivity of central auditory neurons to frequency modulated (FM) sound is often characterized based on spectro-temporal receptive field (STRF), which is generated by spike-trigger averaging a random stimulus. Due to the inherent property of time variability in neural response, this method erroneously represents the response jitter as stimulus jitter in the STRF. To reveal the trigger features more clearly, we have implemented a method that minimizes this error. Neural spikes from the brainstem of urethane-anesthetized rats were first recorded in response to two sets of FM stimuli: (a) a random FM tone for the generation of STRF and (b) a family of linear FM ramps for the determination of FM 'trigger point'. Based on the first dataset, STRFs were generated using spike-trigger averaging. Individual modulating waveforms were then matched with respect to their mean waveform at time-windows of a systematically varied length. A stable or optimal variance time profile was found at a particular window length. At this optimal window length, we performed delay adjustments. A marked sharpening in the FM bands in the STRF was found. Results were consistent with the FM 'trigger point' as estimated by the linear FM ramps. We concluded that the present approach of adjusting response jitter was effective in delineating FM trigger features in the STRF.

3D localization of clustered microcalcifications using cranio-caudal and medio-lateral oblique views.

This paper presents a 3D localization method to register clustered microcalcifications on mammograms from cranio-caudal (CC) and medio-lateral oblique (MLO) views. The method consists of three major components: registration of clustered microcalcifications in CC and MLO views, 3D localization of clustered microcalcifications and 3D visualization of clustered microcalcifications. The registration is performed based on three features, gradient, energy and local entropy codes that are independent of spatial locations of microcalcifications in two different views and are prioritized by discriminability in a binary decision tree. The 3D localization is determined by a sequence of coordinate corrections of calcified pixels using the breast nipple as a controlling point. Finally, the 3D visualization implements a virtual reality modeling language viewer (VRMLV) to view the exact location of the lesion as a guide for needle biopsy. In order to validate our proposed 3D localization system, a set of breast lesions, which appear both in mammograms and in MR Images is used for experiments where the depth of clustered microcalcifications can be verified by the MR images.

Detection of spectral signatures in multispectral MR images for classification.

This paper presents a new spectral signature detection approach to magnetic resonance (MR) image classification. It is called constrained energy minimization (CEM) method, which is derived from the minimum variance distortionless response in passive sensor array processing. It considers a bank of spectral channels as an array of sensors where each spectral channel represents a sensor and object spectral signature in multispectral MR images are viewed as signals impinging upon the array. The strength of the CEM lies on its ability in detection of spectral signatures of interest without knowing image background. The detected spectral signatures are then used for classification. The CEM makes use of a finite impulse response (FIR) filter to linearly constrain a desired object while minimizing interfering effects caused by other unknown signal sources. Unlike most spatial-based classification techniques, the proposed CEM takes advantage of spectral characteristics to achieve object detection and classification. A series of experiments is conducted and compared with the commonly used c-means method for performance evaluation. The results show that the CEM method is a promising and effective spectral technique for MR image classification.

Classification of clustered microcalcifications using a Shape Cognitron neural network.

A new shape recognition-based neural network built with universal feature planes, called Shape Cognitron (S-Cognitron) is introduced to classify clustered microcalcifications. The architecture of S-Cognitron consists of two modules and an extra layer, called 3D figure layer lies in between. The first module contains a shape orientation layer, built with 20 cell planes of low level universal shape features to convert first-order shape orientations into numeric values, and a complex layer, to extract second-order shape features. The 3D figure layer is a feature extract-display layer that extracts the shape curvatures of an input pattern and displays them as a 3D figure. It is then followed by a second module made up of a feature formation layer and a probabilistic neural network-based classification layer. The system is evaluated by using Nijmegen mammogram database and experimental results show that sensitivity and specificity can reach 86.1 and 74.1%, respectively.

Identifying multiple abdominal organs from CT image series using a multimodule contextual neural network and spatial fuzzy rules.

Identifying abdominal organs is one of the essential steps in visualizing organ structure to assist in teaching, clinical training, diagnosis, and medical image retrieval. However, due to partial volume effects, gray-level similarities of adjacent organs, contrast media affect, and the relatively high variations of organ position and shape, automatically identifying abdominal organs has always been a high challenging task. To conquer these difficulties, this paper proposes combining a multimodule contextual neural network and spatial fuzzy rules and fuzzy descriptors for automatically identifying abdominal organs from a series of CT image slices. The multimodule contextual neural network segments each image slice through a divide-and-conquer concept, embedded within multiple neural network modules, where the results obtained from each module are forwarded to other modules for integration, in which contextual constraints are enforced. With this approach, the difficulties arising from partial volume effects, gray-level similarities of adjacent organs, and contrast media affect can be reduced to the extreme. To address the issue of high variations in organ position and shape, spatial fuzzy rules and fuzzy descriptors are adopted, along with a contour modification scheme implementing consecutive organ region overlap constraints. This approach has been tested on 40 sets of abdominal CT images, where each set consists of about 40 image slices. We have found that 99% of the organ regions in the test images are correctly identified as its belonging organs, implying the high promise of the proposed method.

Responses of central auditory neurons modeled with finite impulse response (FIR) neural networks.

To simulate central auditory responses to complex sounds, a computational model was implemented. It consists of a multi-scale classification process, and an artificial neural network composed of two modules of finite impulse response (FIR) neural networks connected to a maximum network. Electrical activities of single auditory neurons were recorded at the rat midbrain in response to a repetitive pseudo-random frequency modulated (FM) sound. The multi-scale classification process divides the training dataset into either strong or weak response using a multiple-scale Gaussian filter that based on response probability. Two modules of FIR neural network are then independently trained to model the two types of responses. This caters for the possible differences in neuronal circuitry and transmission delay. Their outputs are connected to a maximum network to generate the final output. After training, we use a different set of FM responses collected from the same neuron to test the performance of the model. Two criteria are adopted for assessment. One measures the matching of the modeled output to the actual output on a point-to-point basis. Another measures the matching of bulk responses between the two. Results show that the proposed model predicts the responses of central auditory neurons satisfactorily.

Reconstruction of medical images under different image acquisition angles.

Compared to object-based registration, feature-based registration is much less complex. However, in order for feature-based registration to work, the two image stacks under consideration must have the same acquisition tilt angle and the same anatomical location - two requirements that are not always fulfilled. In this paper, we propose a technique that reconstructs two sets of medical images acquired with different acquisition angles and anatomical cross sections into one set of images of identical scanning orientation and positions. The space correlation information among the two image stacks is first extracted and is used to correct the tilt angle and anatomical position differences in the image stacks. Satisfactory reconstruction results were presented to prove our points.

Gait and balance analysis for patients with Alzheimer's disease using an inertial-sensor-based wearable instrument.

Despite patients with Alzheimer's disease (AD) were reported of revealing gait disorders and balance problems, there is still lack of objective quantitative measurement of gait patterns and balance capability of AD patients. Based on an inertial-sensor-based wearable device, this paper develops gait and balance analyzing algorithms to obtain quantitative measurements and explores the essential indicators from the measurements for AD diagnosis. The gait analyzing algorithm is composed of stride detection followed by gait cycle decomposition so that gait parameters are developed from the decomposed gait details. On the other hand, the balance is measured by the sway speed in anterior-posterior (AP) and medial-lateral (ML) directions of the projection path of body's center of mass (COM). These devised gait and balance parameters were explored on twenty-one AD patients and fifty healthy controls (HCs). Special evaluation procedure including single-task and dual-task walking experiments for observing the cognitive function and attention is also devised for the comparison of AD and HC groups. Experimental results show that the wearable instrument with the designed gait and balance analyzing system is a promising tool for automatically analyzing gait information and balance ability, serving as assistant indicators for early diagnosis of AD.