Cancerous and Non-Cancerous MRI Classification Using Dual DCNN Approach.

Brain cancer is a life-threatening disease requiring close attention. Early and accurate diagnosis using non-invasive medical imaging is critical for successful treatment and patient survival. However, manual diagnosis by radiologist experts is time-consuming and has limitations in processing large datasets efficiently. Therefore, efficient systems capable of analyzing vast amounts of medical data for early tumor detection are urgently needed. Deep learning (DL) with deep convolutional neural networks (DCNNs) emerges as a promising tool for understanding diseases like brain cancer through medical imaging modalities, especially MRI, which provides detailed soft tissue contrast for visualizing tumors and organs. DL techniques have become more and more popular in current research on brain tumor detection. Unlike traditional machine learning methods requiring manual feature extraction, DL models are adept at handling complex data like MRIs and excel in classification tasks, making them well-suited for medical image analysis applications. This study presents a novel Dual DCNN model that can accurately classify cancerous and non-cancerous MRI samples. Our Dual DCNN model uses two well-performed DL models, i.e., inceptionV3 and denseNet121. Features are extracted from these models by appending a global max pooling layer. The extracted features are then utilized to train the model with the addition of five fully connected layers and finally accurately classify MRI samples as cancerous or non-cancerous. The fully connected layers are retrained to learn the extracted features for better accuracy. The technique achieves 99%, 99%, 98%, and 99% of accuracy, precision, recall, and f1-scores, respectively. Furthermore, this study compares the Dual DCNN's performance against various well-known DL models, including DenseNet121, InceptionV3, ResNet architectures, EfficientNetB2, SqueezeNet, VGG16, AlexNet, and LeNet-5, with different learning rates. This study indicates that our proposed approach outperforms these established models in terms of performance.

Aligning Small Datasets Using Domain Adversarial Learning: Applications in Automated in Vivo Oral Cancer Diagnosis.

Deep learning approaches for medical image analysis are limited by small data set size due to factors such as patient privacy and difficulties in obtaining expert labelling for each image. In medical imaging system development pipelines, phases for system development and classification algorithms often overlap with data collection, creating small disjoint data sets collected at numerous locations with differing protocols. In this setting, merging data from different data collection centers increases the amount of training data. However, a direct combination of datasets will likely fail due to domain shifts between imaging centers. In contrast to previous approaches that focus on a single data set, we add a domain adaptation module to a neural network and train using multiple data sets. Our approach encourages domain invariance between two multispectral autofluorescence imaging (maFLIM) data sets of in vivo oral lesions collected with an imaging system currently in development. The two data sets have differences in the sub-populations imaged and in the calibration procedures used during data collection. We mitigate these differences using a gradient reversal layer and domain classifier. Our final model trained with two data sets substantially increases performance, including a significant increase in specificity. We also achieve a significant increase in average performance over the best baseline model train with two domains (p = 0.0341). Our approach lays the foundation for faster development of computer-aided diagnostic systems and presents a feasible approach for creating a robust classifier that aligns images from multiple data centers in the presence of domain shifts.

Clinical label-free endoscopic imaging of biochemical and metabolic autofluorescence biomarkers of benign, precancerous, and cancerous oral lesions.

Early detection is critical for improving the survival rate and quality of life of oral cancer patients; unfortunately, dysplastic and early-stage cancerous oral lesions are often difficult to distinguish from oral benign lesions during standard clinical oral examination. Therefore, there is a critical need for novel clinical technologies that would enable reliable oral cancer screening. The autofluorescence properties of the oral epithelial tissue provide quantitative information about morphological, biochemical, and metabolic tissue and cellular alterations accompanying carcinogenesis. This study aimed to identify novel biochemical and metabolic autofluorescence biomarkers of oral dysplasia and cancer that could be clinically imaged using novel multispectral autofluorescence lifetime imaging (maFLIM) endoscopy technologies. In vivo maFLIM clinical endoscopic images of benign, precancerous, and cancerous lesions from 67 patients were acquired using a novel maFLIM endoscope. Widefield maFLIM feature maps were generated, and statistical analyses were applied to identify maFLIM features providing contrast between dysplastic/cancerous vs. benign oral lesions. A total of 14 spectral and time-resolved maFLIM features were found to provide contrast between dysplastic/cancerous vs. benign oral lesions, representing novel biochemical and metabolic autofluorescence biomarkers of oral epithelial dysplasia and cancer. To the best of our knowledge, this is the first demonstration of clinical widefield maFLIM endoscopic imaging of novel biochemical and metabolic autofluorescence biomarkers of oral dysplasia and cancer, supporting the potential of maFLIM endoscopy for early detection of oral cancer.

End-to-End Neural Network for Feature Extraction and Cancer Diagnosis of In Vivo Fluorescence Lifetime Images of Oral Lesions.

In contrast to previous studies that focused on classical machine learning algorithms and hand-crafted features, we present an end-to-end neural network classification method able to accommodate lesion heterogeneity for improved oral cancer diagnosis using multispectral autofluorescence lifetime imaging (maFLIM) endoscopy. Our method uses an autoencoder framework jointly trained with a classifier designed to handle overfitting problems with reduced databases, which is often the case in healthcare applications. The autoencoder guides the feature extraction process through the reconstruction loss and enables the potential use of unsupervised data for domain adaptation and improved generalization. The classifier ensures the features extracted are task-specific, providing discriminative information for the classification task. The data-driven feature extraction method automatically generates task-specific features directly from fluorescence decays, eliminating the need for iterative signal reconstruction. We validate our proposed neural network method against support vector machine (SVM) baselines, with our method showing a 6.5%-8.3% increase in sensitivity. Our results show that neural networks that implement data-driven feature extraction provide superior results and enable the capacity needed to target specific issues, such as inter-patient variability and the heterogeneity of oral lesions.Clinical relevance- We improve standard classification algorithms for in vivo diagnosis of oral cancer lesions from maFLIm for clinical use in cancer screening, reducing unnecessary biopsies and facilitating early detection of oral cancer.

Machine-Learning Assisted Discrimination of Precancerous and Cancerous from Healthy Oral Tissue Based on Multispectral Autofluorescence Lifetime Imaging Endoscopy.

Multispectral autofluorescence lifetime imaging (maFLIM) can be used to clinically image a plurality of metabolic and biochemical autofluorescence biomarkers of oral epithelial dysplasia and cancer. This study tested the hypothesis that maFLIM-derived autofluorescence biomarkers can be used in machine-learning (ML) models to discriminate dysplastic and cancerous from healthy oral tissue. Clinical widefield maFLIM endoscopy imaging of cancerous and dysplastic oral lesions was performed at two clinical centers. Endoscopic maFLIM images from 34 patients acquired at one of the clinical centers were used to optimize ML models for automated discrimination of dysplastic and cancerous from healthy oral tissue. A computer-aided detection system was developed and applied to a set of endoscopic maFLIM images from 23 patients acquired at the other clinical center, and its performance was quantified in terms of the area under the receiver operating characteristic curve (ROC-AUC). Discrimination of dysplastic and cancerous from healthy oral tissue was achieved with an ROC-AUC of 0.81. This study demonstrates the capabilities of widefield maFLIM endoscopy to clinically image autofluorescence biomarkers that can be used in ML models to discriminate dysplastic and cancerous from healthy oral tissue. Widefield maFLIM endoscopy thus holds potential for automated in situ detection of oral dysplasia and cancer.

Using MRI to quantify skeletal muscle pathology in Duchenne muscular dystrophy: A systematic mapping review.

There is a great demand for accurate non-invasive measures to better define the natural history of disease progression or treatment outcome in Duchenne muscular dystrophy (DMD) and to facilitate the inclusion of a large range of participants in DMD clinical trials. This review aims to investigate which MRI sequences and analysis methods have been used and to identify future needs. Medline, Embase, Scopus, Web of Science, Inspec, and Compendex databases were searched up to 2 November 2019, using keywords "magnetic resonance imaging" and "Duchenne muscular dystrophy." The review showed the trend of using T1w and T2w MRI images for semi-qualitative inspection of structural alterations of DMD muscle using a diversity of grading scales, with increasing use of T2map, Dixon, and MR spectroscopy (MRS). High-field (>3T) MRI dominated the studies with animal models. The quantitative MRI techniques have allowed a more precise estimation of local or generalized disease severity. Longitudinal studies assessing the effect of an intervention have also become more prominent, in both clinical and animal model subjects. Quality assessment of the included longitudinal studies was performed using the Newcastle-Ottawa Quality Assessment Scale adapted to comprise bias in selection, comparability, exposure, and outcome. Additional large clinical trials are needed to consolidate research using MRI as a biomarker in DMD and to validate findings against established gold standards. This future work should use a multiparametric and quantitative MRI acquisition protocol, assess the repeatability of measurements, and correlate findings to histologic parameters.

MR Image-Based Attenuation Correction of Brain PET Imaging: Review of Literature on Machine Learning Approaches for Segmentation.

Recent emerging hybrid technology of positron emission tomography/magnetic resonance (PET/MR) imaging has generated a great need for an accurate MR image-based PET attenuation correction. MR image segmentation, as a robust and simple method for PET attenuation correction, has been clinically adopted in commercial PET/MR scanners. The general approach in this method is to segment the MR image into different tissue types, each assigned an attenuation constant as in an X-ray CT image. Machine learning techniques such as clustering, classification and deep networks are extensively used for brain MR image segmentation. However, only limited work has been reported on using deep learning in brain PET attenuation correction. In addition, there is a lack of clinical evaluation of machine learning methods in this application. The aim of this review is to study the use of machine learning methods for MR image segmentation and its application in attenuation correction for PET brain imaging. Furthermore, challenges and future opportunities in MR image-based PET attenuation correction are discussed.

Clinical label-free biochemical and metabolic fluorescence lifetime endoscopic imaging of precancerous and cancerous oral lesions.

INTRODUCTION: Incomplete head and neck cancer resection occurs in up to 85% of cases, leading to increased odds of local recurrence and regional metastases; thus, image-guided surgical tools for accurate, in situ and fast detection of positive margins during head and neck cancer resection surgery are urgently needed. Oral epithelial dysplasia and cancer development is accompanied by morphological, biochemical, and metabolic tissue and cellular alterations that can modulate the autofluorescence properties of the oral epithelial tissue. OBJECTIVE: This study aimed to test the hypothesis that autofluorescence biomarkers of oral precancer and cancer can be clinically imaged and quantified by means of multispectral fluorescence lifetime imaging (FLIM) endoscopy. METHODS: Multispectral autofluorescence lifetime images of precancerous and cancerous lesions from 39 patients were imaged in vivo using a novel multispectral FLIM endoscope and processed to generate widefield maps of biochemical and metabolic autofluorescence biomarkers of oral precancer and cancer. RESULTS: Statistical analyses applied to the quantified multispectral FLIM endoscopy based autofluorescence biomarkers indicated their potential to provide contrast between precancerous/cancerous vs. healthy oral epithelial tissue. CONCLUSION: To the best of our knowledge, this study represents the first demonstration of label-free biochemical and metabolic clinical imaging of precancerous and cancerous oral lesions by means of widefield multispectral autofluorescence lifetime endoscopy. Future studies will focus on demonstrating the capabilities of endogenous multispectral FLIM endoscopy as an image-guided surgical tool for positive margin detection during head and neck cancer resection surgery.

Predicting long-term type 2 diabetes with support vector machine using oral glucose tolerance test.

Diabetes is a large healthcare burden worldwide. There is substantial evidence that lifestyle modifications and drug intervention can prevent diabetes, therefore, an early identification of high risk individuals is important to design targeted prevention strategies. In this paper, we present an automatic tool that uses machine learning techniques to predict the development of type 2 diabetes mellitus (T2DM). Data generated from an oral glucose tolerance test (OGTT) was used to develop a predictive model based on the support vector machine (SVM). We trained and validated the models using the OGTT and demographic data of 1,492 healthy individuals collected during the San Antonio Heart Study. This study collected plasma glucose and insulin concentrations before glucose intake and at three time-points thereafter (30, 60 and 120 min). Furthermore, personal information such as age, ethnicity and body-mass index was also a part of the data-set. Using 11 OGTT measurements, we have deduced 61 features, which are then assigned a rank and the top ten features are shortlisted using minimum redundancy maximum relevance feature selection algorithm. All possible combinations of the 10 best ranked features were used to generate SVM based prediction models. This research shows that an individual's plasma glucose levels, and the information derived therefrom have the strongest predictive performance for the future development of T2DM. Significantly, insulin and demographic features do not provide additional performance improvement for diabetes prediction. The results of this work identify the parsimonious clinical data needed to be collected for an efficient prediction of T2DM. Our approach shows an average accuracy of 96.80% and a sensitivity of 80.09% obtained on a holdout set.

Muscle percentage index as a marker of disease severity in golden retriever muscular dystrophy.

INTRODUCTION: Golden retriever muscular dystrophy (GRMD) is a spontaneous X-linked canine model of Duchenne muscular dystrophy that resembles the human condition. Muscle percentage index (MPI) is proposed as an imaging biomarker of disease severity in GRMD. METHODS: To assess MPI, we used MRI data acquired from nine GRMD samples using a 4.7 T small-bore scanner. A machine learning approach was used with eight raw quantitative mapping of MRI data images (T1m, T2m, two Dixon maps, and four diffusion tensor imaging maps), three types of texture descriptors (local binary pattern, gray-level co-occurrence matrix, gray-level run-length matrix), and a gradient descriptor (histogram of oriented gradients). RESULTS: The confusion matrix, averaged over all samples, showed 93.5% of muscle pixels classified correctly. The classification, optimized in a leave-one-out cross-validation, provided an average accuracy of 80% with a discrepancy in overestimation for young (8%) and old (20%) dogs. DISCUSSION: MPI could be useful for quantifying GRMD severity, but careful interpretation is needed for severe cases.

Accelerated positive contrast MRI of interventional devices using parallel compressed sensing imaging.

Susceptibility-based magnetic resonance imaging (MRI) method can image small MR-compatible devices with positive contrast. However, the relatively long data acquisition time required by the method hinders its practical applications. This study presents a parallel compressive sensing technique with a modified fast spin echo to accelerate data acquisition for the susceptibility-based positive contrast MRI. The method integrates the generalized autocalibrating partially parallel acquisitions and the compressive sensing techniques in the reconstruction algorithm. MR imaging data acquired from several phantoms containing interventional devices such as biopsy needles, stent, and brachytherapy seeds, used for validating the proposed technique. The results show that it can speed up data acquisition by a factor of about five while preserving the quality of the positive contrast images.

New Similarity Metric for Registration of MRI to Histology: Golden Retriever Muscular Dystrophy Imaging.

OBJECTIVE: Histology is often used as a gold standard to evaluate noninvasive imaging modalities such as a magnetic resonance imaging (MRI). Spatial correspondence between histology and MRI is a critical step in quantitative evaluation of skeletal muscle in golden retriever muscular dystrophy (GRMD). Registration becomes technically challenging due to nonorthogonal histology section orientation, section distortion, and the different image contrast and resolution. METHODS: This study describes a three-step procedure to register histology images with multiparametric MRI, i.e., interactive slice localization controlled by a three-dimensional mouse, followed by an affine transformation refinement, and a B-spline deformable registration using a new similarity metric. This metric combines mutual information and gradient information. RESULTS: The methodology was verified using ex vivo high-resolution multiparametric MRI with a resolution of 117.19 µm (i.e., T1-weighted and T2-weighted MRI images) and trichrome stained histology images acquired from the pectineus muscles of ten dogs (nine GRMD and one healthy control). The proposed registration method yielded a root mean squares (RMS) error of 148.83 ± 34.96 µm averaged for ten muscle samples based on landmark points validated by five observers. The best RMS error averaged for ten muscles, was 128.48 ± 25.39 µm. CONCLUSION: The established correspondence between histology and in vivo MRI enables accurate extraction of MRI characteristics for histologically confirmed regions (e.g., muscle, fibrosis, and fat). SIGNIFICANCE: The proposed methodology allows creation of a database of spatially registered multiparametric MRI and histology. This database will facilitate accurate monitoring of disease progression and assess treatment effects noninvasively.

Texture as an imaging biomarker for disease severity in golden retriever muscular dystrophy.

INTRODUCTION: Golden retriever muscular dystrophy (GRMD), an X-linked recessive disorder, causes similar phenotypic features to Duchenne muscular dystrophy (DMD). There is currently a need for a quantitative and reproducible monitoring of disease progression for GRMD and DMD. METHODS: To assess severity in the GRMD, we analyzed texture features extracted from multi-parametric MRI (T1w, T2w, T1m, T2m, and Dixon images) using 5 feature extraction methods and classified using support vector machines. RESULTS: A single feature from qualitative images can provide 89% maximal accuracy. Furthermore, 2 features from T1w, T2m, or Dixon images provided highest accuracy. When considering a tradeoff between scan-time and computational complexity, T2m images provided good accuracy at a lower acquisition and processing time and effort. CONCLUSIONS: The combination of MRI texture features improved the classification accuracy for assessment of disease progression in GRMD with evaluation of the heterogenous nature of skeletal muscles as reflection of the histopathological changes. Muscle Nerve 59:380-386, 2019.

Implementation of High-Performance Correlation and Mapping Engine for Rapid Generation of Brain Connectivity Networks from Big fMRI Data.

With the emergence of the dynamic functional connectivity analysis, and the studies relying on real-time neurological feedback, the need for rapid processing methods becomes even more critical. Seed-based Correlation Analysis (SCA) of fMRI data has been used to create brain connectivity networks. With close to a million voxels in a fMRI dataset, the number of calculations involved in SCA becomes high. This work aims to demonstrate a new approach which produces high-resolution brain connectivity maps rapidly. The results show that HPCME with four FPGAs can improve the SCA processing speed by a factor of 40 or more over that of a PC workstation with a multicore CPU.

MARBLES - Metal Artifact Based Landmark Enhanced Susceptibility Weighted Imaging For Interventional Device Localization In MRI.

Susceptibility Weighted Imaging (SWI) is a method extensively studied for its application to improve contrast in MR imaging modality. The method enhances the visualization of magnetically susceptible content such as iron, calcium and zinc in the tissues by using the susceptibility differences in tissues to generate a unique image contrast. In this study, we propose an SWI based approach to improve the visualization of interventional devices in MRI data. Results obtained from two datasets (biopsy needle and brachytherapy seeds), indicate SWI to be suitable for visualization of the interventional devices, while also being computationally faster when compared with quantitative susceptibility mapping (QSM).

Three-dimensional self-gated cardiac MR imaging for the evaluation of myocardial infarction in mouse model on a 3T clinical MR system.

PURPOSE: To develop and assess a three-dimensional (3D) self-gated technique for the evaluation of myocardial infarction (MI) in mouse model without the use of external electrocardiogram (ECG) trigger and respiratory motion sensor on a 3T clinical MR system. METHODS: A 3D T1-weighted GRE sequence with stack-of-stars sampling trajectories was developed and performed on six mice with MIs that were injected with a gadolinium-based contrast agent at a 3T clinical MR system. Respiratory and cardiac self-gating signals were derived from the Cartesian mapping of the k-space center along the partition encoding direction by bandpass filtering in image domain. The data were then realigned according to the predetermined self-gating signals for the following image reconstruction. In order to accelerate the data acquisition, image reconstruction was based on compressed sensing (CS) theory by exploiting temporal sparsity of the reconstructed images. In addition, images were also reconstructed from the same realigned data by conventional regridding method for demonstrating the advantageous of the proposed reconstruction method. Furthermore, the accuracy of detecting MI by the proposed method was assessed using histological analysis as the standard reference. Linear regression and Bland-Altman analysis were used to assess the agreement between the proposed method and the histological analysis. RESULTS: Compared to the conventional regridding method, the proposed CS method reconstructed images with much less streaking artifact, as well as a better contrast-to-noise ratio (CNR) between the blood and myocardium (4.1 ± 2.1 vs. 2.9 ± 1.1, p = 0.031). Linear regression and Bland-Altman analysis demonstrated that excellent correlation was obtained between infarct sizes derived from the proposed method and histology analysis. CONCLUSION: A 3D T1-weighted self-gating technique for mouse cardiac imaging was developed, which has potential for accurately evaluating MIs in mice at 3T clinical MR system without the use of external ECG trigger and respiratory motion sensor.

Parallel compressive sensing in a hybrid space: Application in interventional MRI.

Recently, susceptibility based positive contrast MRI technique emerged as an effective method of visualizing the small MR compatible devices, such as brachytherapy seeds. One of the challenges associated with this method is the long scan time. In this work, we present an accelerated susceptibility based positive contrast MR imaging method, in which the susceptibility map can be generated from an under-sampled data. We use a combination of parallel imaging (GRAPPA) and compressive sensing (CS) technique in a hybrid k-space. The results in brachytherapy seeds imaging show that 3-D high-quality images can be achieved at an acceleration factor up to 4, which can decrease the scan time from 4 minutes to 1.4 minutes.

Black-blood T2* mapping with delay alternating with nutation for tailored excitation.

PURPOSE: To develop a black-blood T2* mapping method using a Delay Alternating with Nutation for Tailored Excitation (DANTE) preparation combined with a multi-echo gradient echo (GRE) readout (DANTE-GRE). MATERIALS AND METHODS: Simulations of the Bloch equation for DANTE-GRE were performed to optimize sequence parameters. After optimization, the sequence was applied to a phantom scan and to neck and lower extremity scans conducted on 12 volunteers at 3T using DANTE-GRE, Motion-Sensitized Driven Equilibrium (MSDE)-GRE, and multi-echo GRE. T2* values were measured using an offset model. Statistical analyses were conducted to compare the T2* values between the three sequences. RESULTS: Simulation results showed that blood suppression can be achieved with various DANTE parameter adjustments. T2* maps acquired by DANTE-GRE were consistent and comparable to those acquired with multi-echo GRE in phantom experiments. In the in vivo experiments, DANTE-GRE was more comparable to multi-echo GRE than MSDE-GRE regarding the measurement of muscle T2* values. CONCLUSION: Due to its high signal intensity retention and effective blood signal suppression, DANTE-GRE allows for robust and accurate T2* quantification, superior to that of MSDE-GRE, while overcoming blood flow artifacts associated with traditional multi-echo GRE.

Accelerated susceptibility-based positive contrast imaging of MR compatible metallic devices based on modified fast spin echo sequences.

This study aims to develop an accelerated susceptibility-based positive contrast MR imaging method for visualizing MR compatible metallic devices. A modified fast spin echo sequence is used to accelerate data acquisition. Each readout gradient in the modified fast spin echo is slightly shifted by a short distance T shift. Phase changes accumulated within T shift are then used to calculate the susceptibility map by using a kernel deconvolution algorithm with a regularized ℓ1 minimization. To evaluate the proposed fast spin echo method, three phantom experiments were conducted and compared to a spin echo based technique and the gold standard CT for visualizing biopsy needles and brachytherapy seeds. Compared to the spin echo based technique, the data sampling speed of the proposed method was faster by 2-4 times while still being able to accurately visualize and identify the location of the biopsy needle and brachytherapy seeds. These results were confirmed by CT images of the same devices. Results also demonstrated that the proposed fast spin echo method can achieve good visualization of the brachytherapy seeds in positive contrast and in different orientations. It is also capable of correctly differentiating brachytherapy seeds from other similar structures on conventional magnitude images.

Compressed sensing MRI reconstruction from 3D multichannel data using GPUs.

PURPOSE: To accelerate iterative reconstructions of compressed sensing (CS) MRI from 3D multichannel data using graphics processing units (GPUs). METHODS: The sparsity of MRI signals and parallel array receivers can reduce the data acquisition requirements. However, iterative CS reconstructions from data acquired using an array system may take a significantly long time, especially for a large number of parallel channels. This paper presents an efficient method for CS-MRI reconstruction from 3D multichannel data using GPUs. In this method, CS reconstructions were simultaneously processed in a channel-by-channel fashion on the GPU, in which the computations of multiple-channel 3D-CS reconstructions are highly parallelized. The final image was then produced by a sum-of-squares method on the central processing unit. Implementation details including algorithm, data/memory management, and parallelization schemes are reported in the paper. RESULTS: Both simulated data and in vivo MRI array data were tested. The results showed that the proposed method can significantly improve the image reconstruction efficiency, typically shortening the runtime by a factor of 30. CONCLUSIONS: Using low-cost GPUs and an efficient algorithm allowed the 3D multislice compressive-sensing reconstruction to be performed in less than 1 s. The rapid reconstructions are expected to help bring high-dimensional, multichannel parallel CS MRI closer to clinical applications. Magn Reson Med 78:2265-2274, 2017. © 2017 International Society for Magnetic Resonance in Medicine.