Automatic Active Lesion Tracking in Multiple Sclerosis Using Unsupervised Machine Learning.

BACKGROUND: Identifying active lesions in magnetic resonance imaging (MRI) is crucial for the diagnosis and treatment planning of multiple sclerosis (MS). Active lesions on MRI are identified following the administration of Gadolinium-based contrast agents (GBCAs). However, recent studies have reported that repeated administration of GBCA results in the accumulation of Gd in tissues. In addition, GBCA administration increases health care costs. Thus, reducing or eliminating GBCA administration for active lesion detection is important for improved patient safety and reduced healthcare costs. Current state-of-the-art methods for identifying active lesions in brain MRI without GBCA administration utilize data-intensive deep learning methods. OBJECTIVE: To implement nonlinear dimensionality reduction (NLDR) methods, locally linear embedding (LLE) and isometric feature mapping (Isomap), which are less data-intensive, for automatically identifying active lesions on brain MRI in MS patients, without the administration of contrast agents. MATERIALS AND METHODS: Fluid-attenuated inversion recovery (FLAIR), T2-weighted, proton density-weighted, and pre- and post-contrast T1-weighted images were included in the multiparametric MRI dataset used in this study. Subtracted pre- and post-contrast T1-weighted images were labeled by experts as active lesions (ground truth). Unsupervised methods, LLE and Isomap, were used to reconstruct multiparametric brain MR images into a single embedded image. Active lesions were identified on the embedded images and compared with ground truth lesions. The performance of NLDR methods was evaluated by calculating the Dice similarity (DS) index between the observed and identified active lesions in embedded images. RESULTS: LLE and Isomap, were applied to 40 MS patients, achieving median DS scores of 0.74 ± 0.1 and 0.78 ± 0.09, respectively, outperforming current state-of-the-art methods. CONCLUSIONS: NLDR methods, Isomap and LLE, are viable options for the identification of active MS lesions on non-contrast images, and potentially could be used as a clinical decision tool.

A Novel Machine-Learning Framework Based on a Hierarchy of Dispute Models for the Identification of Fish Species Using Multi-Mode Spectroscopy.

Seafood mislabeling rates of approximately 20% have been reported globally. Traditional methods for fish species identification, such as DNA analysis and polymerase chain reaction (PCR), are expensive and time-consuming, and require skilled technicians and specialized equipment. The combination of spectroscopy and machine learning presents a promising approach to overcome these challenges. In our study, we took a comprehensive approach by considering a total of 43 different fish species and employing three modes of spectroscopy: fluorescence (Fluor), and reflectance in the visible near-infrared (VNIR) and short-wave near-infrared (SWIR). To achieve higher accuracies, we developed a novel machine-learning framework, where groups of similar fish types were identified and specialized classifiers were trained for each group. The incorporation of global (single artificial intelligence for all species) and dispute classification models created a hierarchical decision process, yielding higher performances. For Fluor, VNIR, and SWIR, accuracies increased from 80%, 75%, and 49% to 83%, 81%, and 58%, respectively. Furthermore, certain species witnessed remarkable performance enhancements of up to 40% in single-mode identification. The fusion of all three spectroscopic modes further boosted the performance of the best single mode, averaged over all species, by 9%. Fish species mislabeling not only poses health-related risks due to contaminants, toxins, and allergens that could be life-threatening, but also gives rise to economic and environmental hazards and loss of nutritional benefits. Our proposed method can detect fish fraud as a real-time alternative to DNA barcoding and other standard methods. The hierarchical system of dispute models proposed in this work is a novel machine-learning tool not limited to this application, and can improve accuracy in any classification problem which contains a large number of classes.

Medical Imaging Applications of Federated Learning.

Since its introduction in 2016, researchers have applied the idea of Federated Learning (FL) to several domains ranging from edge computing to banking. The technique's inherent security benefits, privacy-preserving capabilities, ease of scalability, and ability to transcend data biases have motivated researchers to use this tool on healthcare datasets. While several reviews exist detailing FL and its applications, this review focuses solely on the different applications of FL to medical imaging datasets, grouping applications by diseases, modality, and/or part of the body. This Systematic Literature review was conducted by querying and consolidating results from ArXiv, IEEE Xplorer, and PubMed. Furthermore, we provide a detailed description of FL architecture, models, descriptions of the performance achieved by FL models, and how results compare with traditional Machine Learning (ML) models. Additionally, we discuss the security benefits, highlighting two primary forms of privacy-preserving techniques, including homomorphic encryption and differential privacy. Finally, we provide some background information and context regarding where the contributions lie. The background information is organized into the following categories: architecture/setup type, data-related topics, security, and learning types. While progress has been made within the field of FL and medical imaging, much room for improvement and understanding remains, with an emphasis on security and data issues remaining the primary concerns for researchers. Therefore, improvements are constantly pushing the field forward. Finally, we highlighted the challenges in deploying FL in medical imaging applications and provided recommendations for future directions.

Automated Lung Ultrasound Pulmonary Disease Quantification Using an Unsupervised Machine Learning Technique for COVID-19.

COVID-19 is an ongoing global health pandemic. Although COVID-19 can be diagnosed with various tests such as PCR, these tests do not establish pulmonary disease burden. Whereas point-of-care lung ultrasound (POCUS) can directly assess the severity of characteristic pulmonary findings of COVID-19, the advantage of using US is that it is inexpensive, portable, and widely available for use in many clinical settings. For automated assessment of pulmonary findings, we have developed an unsupervised learning technique termed the calculated lung ultrasound (CLU) index. The CLU can quantify various types of lung findings, such as A or B lines, consolidations, and pleural effusions, and it uses these findings to calculate a CLU index score, which is a quantitative measure of pulmonary disease burden. This is accomplished using an unsupervised, patient-specific approach that does not require training on a large dataset. The CLU was tested on 52 lung ultrasound examinations from several institutions. CLU demonstrated excellent concordance with radiologist findings in different pulmonary disease states. Given the global nature of COVID-19, the CLU would be useful for sonographers and physicians in resource-strapped areas with limited ultrasound training and diagnostic capacities for more accurate assessment of pulmonary status.

Rapid Assessment of Fish Freshness for Multiple Supply-Chain Nodes Using Multi-Mode Spectroscopy and Fusion-Based Artificial Intelligence.

This study is directed towards developing a fast, non-destructive, and easy-to-use handheld multimode spectroscopic system for fish quality assessment. We apply data fusion of visible near infra-red (VIS-NIR) and short wave infra-red (SWIR) reflectance and fluorescence (FL) spectroscopy data features to classify fish from fresh to spoiled condition. Farmed Atlantic and wild coho and chinook salmon and sablefish fillets were measured. Three hundred measurement points on each of four fillets were taken every two days over 14 days for a total of 8400 measurements for each spectral mode. Multiple machine learning techniques including principal component analysis, self-organized maps, linear and quadratic discriminant analyses, k-nearest neighbors, random forest, support vector machine, and linear regression, as well as ensemble and majority voting methods, were used to explore spectroscopy data measured on fillets and to train classification models to predict freshness. Our results show that multi-mode spectroscopy achieves 95% accuracy, improving the accuracies of the FL, VIS-NIR and SWIR single-mode spectroscopies by 26, 10 and 9%, respectively. We conclude that multi-mode spectroscopy and data fusion analysis has the potential to accurately assess freshness and predict shelf life for fish fillets and recommend this study be expanded to a larger number of species in the future.

A Forehead Wearable Sensor for the Objective Measurement of Chronic Pain.

Chronic pain impacts one in five Americans and is difficult to manage, costing ~USD 600 billion annually. The subjective experience of pain is a complex processing of central nervous system input. Recent advances in magnetic resonance imaging revealed the prefrontal cortex as vital to the perception of pain and that changes in the cerebral hemodynamics can be used to detect painful sensations. Current pain monitoring is dependent on the subjective rating provided by patients and is limited to a single time point. We have developed a biomarker for the objective, real-time and continuous chronic pain assessment using proprietary algorithms termed ROPA and cerebral optical spectrometry. Using a forehead sensor, the cerebral optical spectrometry data were collected in two clinical sites from 41 patients (19 and 22, respectively, from sites 1 and 2), who elected to receive an epidural steroid injection for the treatment of chronic pain. Patients rated their pain on a numeric rating scale, ranging from 0-10, which were used to validate the ROPA objective pain scoring. Multiple time points, including pre- and post-procedure were recorded. The steroid injection was performed per standard medical practice. There was a significant correlation between the patient's reported numeric rating scale and ROPA, for both clinical sites (Overall ~0.81). Holding the subjective pain ratings on a numeric rating scale as ground truth, we determined that the area under the receiver operator curves for both sites revealed at least good (AUC: 64%) to excellent (AUC > 98%) distinctions between clinically meaningful pain severity differentiations (no/mild/moderate/severe). The objective measure of chronic pain (ROPA) determined using cerebral optical spectrometry significantly correlated with the subjective pain scores reported by the subjects. This technology may provide a useful method of detection for the objective and continuous monitoring and treatment of patients with chronic pain, particularly in clinical circumstances where direct assessment is not available, or to complement the patient-reported pain scores.

Combining deep learning and fluorescence imaging to automatically identify fecal contamination on meat carcasses.

Food safety and foodborne diseases are significant global public health concerns. Meat and poultry carcasses can be contaminated by pathogens like E. coli and salmonella, by contact with animal fecal matter and ingesta during slaughter and processing. Since fecal matter and ingesta can host these pathogens, detection, and excision of contaminated regions on meat surfaces is crucial. Fluorescence imaging has proven its potential for the detection of fecal residue but requires expertise to interpret. In order to be used by meat cutters without special training, automated detection is needed. This study used fluorescence imaging and deep learning algorithms to automatically detect and segment areas of fecal matter in carcass images using EfficientNet-B0 to determine which meat surface images showed fecal contamination and then U-Net to precisely segment the areas of contamination. The EfficientNet-B0 model achieved a 97.32% accuracy (precision 97.66%, recall 97.06%, specificity 97.59%, F-score 97.35%) for discriminating clean and contaminated areas on carcasses. U-Net segmented areas with fecal residue with an intersection over union (IoU) score of 89.34% (precision 92.95%, recall 95.84%, specificity 99.79%, F-score 94.37%, and AUC 99.54%). These results demonstrate that the combination of deep learning and fluorescence imaging techniques can improve food safety assurance by allowing the industry to use CSI-D fluorescence imaging to train employees in trimming carcasses as part of their Hazard Analysis Critical Control Point zero-tolerance plan.

FootAssure: A multimodal, in-home wound detection device for diabetic peripheral neuropathy.

Currently, there is no single technology capable of assessing all the multitude of factors associated with peripheral complications of diabetic neuropathy. In this work, a multimodal wound detection system is proposed to help facilitate in-home examinations, utilizing a combination of thermal, multi-spectral 3D imaging modalities. The proposed system is capable of the 3D surface rendering of the foot and would overlay thermal, blood oxygenation, besides other skin health information to aid with foot health monitoring. Examples of biomarkers include pre-ulcer formation, blood circulation, temperature change, oxygenation, swelling, blisters/ulcer formation and healing, and toe health.

Handheld Multispectral Fluorescence Imaging System to Detect and Disinfect Surface Contamination.

Contamination inspection is an ongoing concern for food distributors, restaurant owners, caterers, and others who handle food. Food contamination must be prevented, and zero tolerance legal requirements and damage to the reputation of institutions or restaurants can be very costly. This paper introduces a new handheld fluorescence-based imaging system that can rapidly detect, disinfect, and document invisible organic residues and biofilms which may host pathogens. The contamination, sanitization inspection, and disinfection (CSI-D) system uses light at two fluorescence excitation wavelengths, ultraviolet C (UVC) at 275 nm and violet at 405 nm, for the detection of organic residues, including saliva and respiratory droplets. The 275 nm light is also utilized to disinfect pathogens commonly found within the contaminated residues. Efficacy testing of the neutralizing effects of the ultraviolet light was conducted for Aspergillus fumigatus, Streptococcus pneumoniae, and the influenza A virus (a fungus, a bacterium, and a virus, respectively, each commonly found in saliva and respiratory droplets). After the exposure to UVC light from the CSI-D, all three pathogens experienced deactivation (> 99.99%) in under ten seconds. Up to five-log reductions have also been shown within 10 s of UVC irradiation from the CSI-D system.

Biosensor vital sign detects multiple sclerosis progression.

OBJECTIVE: To determine whether a small, wearable multisensor device can discriminate between progressive versus relapsing multiple sclerosis (MS) and capture limb progression over a short interval, using finger and foot tap data. METHODS: Patients with MS were followed prospectively during routine clinic visits approximately every 6 months. At each visit, participants performed finger and foot taps wearing the MYO-band, which includes accelerometer, gyroscope, and surface electromyogram sensors. Metrics of within-patient limb progression were created by combining the change in signal waveform features over time. The resulting upper (UE) and lower (LE) extremity metrics' discrimination of progressive versus relapsing MS were evaluated with calculation of AUROC. Comparisons with Expanded Disability Status Scale (EDSS) scores were made with Pearson correlation. RESULTS: Participants included 53 relapsing and 15 progressive MS (72% female, baseline mean age 48 years, median disease duration 11 years, median EDSS 2.5, median 10 months follow-up). The final summary metrics differentiated relapsing from secondary progressive MS with AUROC UE 0.93 and LE 0.96. The metrics were associated with baseline EDSS (UE P = 0.0003, LE P = 0.0007). While most had no change in EDSS during the short follow-up, several had evidence of progression by the multisensor metrics. INTERPRETATION: Within a short follow-up interval, this novel multisensor algorithm distinguished progressive from relapsing MS and captured changes in limb function. Inexpensive, noninvasive and easy to use, this novel outcome is readily adaptable to clinical practice and trials as a MS vital sign. This approach also holds promise to monitor limb dysfunction in other neurological diseases.

Spatial Characterization of Tumor Perfusion Properties from 3D DCE-US Perfusion Maps are Early Predictors of Cancer Treatment Response.

There is a need for noninvasive repeatable biomarkers to detect early cancer treatment response and spare non-responders unnecessary morbidities and costs. Here, we introduce three-dimensional (3D) dynamic contrast enhanced ultrasound (DCE-US) perfusion map characterization as inexpensive, bedside and longitudinal indicator of tumor perfusion for prediction of vascular changes and therapy response. More specifically, we developed computational tools to generate perfusion maps in 3D of tumor blood flow, and identified repeatable quantitative features to use in machine-learning models to capture subtle multi-parametric perfusion properties, including heterogeneity. Models were developed and trained in mice data and tested in a separate mouse cohort, as well as early validation clinical data consisting of patients receiving therapy for liver metastases. Models had excellent (ROC-AUC > 0.9) prediction of response in pre-clinical data, as well as proof-of-concept clinical data. Significant correlations with histological assessments of tumor vasculature were noted (Spearman R > 0.70) in pre-clinical data. Our approach can identify responders based on early perfusion changes, using perfusion properties correlated to gold-standard vascular properties.

Novel MS vital sign: multi-sensor captures upper and lower limb dysfunction.

OBJECTIVE: To create a novel neurological vital sign and reliably capture MS-related limb disability in less than 5 min. METHODS: Consecutive patients meeting the 2010 MS diagnostic criteria and healthy controls were offered enrollment. Participants completed finger and foot taps wearing the MYO-band© (accelerometer, gyroscope, and surface electromyogram sensors). Signal processing was performed to extract spatiotemporal features from raw sensor data. Intraclass correlation coefficients (ICC) assessed intertest reproducibility. Spearman correlation and multivariable regression methods compared extracted features to physician- and patient-reported disability outcomes. Partial least squares regression identified the most informative extracted textural features. RESULTS: Baseline data for 117 participants with MS (EDSS 1.0-7.0) and 30 healthy controls were analyzed. ICCs for final selected features ranged from 0.80 to 0.87. Time-based features distinguished cases from controls (P = 0.002). The most informative combination of extracted features from all three sensors strongly correlated with physician EDSS (finger taps rs  = 0.77, P < 0.0001; foot taps rs  = 0.82, P < 0.0001) and had equally strong associations with patient-reported outcomes (WHODAS, finger taps rs  = 0.82, P < 0.0001; foot taps rs  = 0.82, P < 0.0001). Associations remained with multivariable modeling adjusted for age and sex. CONCLUSIONS: Extracted features from the multi-sensor demonstrate striking correlations with gold standard outcomes. Ideal for future generalizability, the assessments take only a few minutes, can be performed by nonclinical personnel, and wearing the band is nondisruptive to routine practice. This novel paradigm holds promise as a new neurological vital sign.

A New Multimodel Machine Learning Framework to Improve Hepatic Fibrosis Grading Using Ultrasound Elastography Systems from Different Vendors.

The purpose of the work described here was to determine if the diagnostic performance of point and 2-D shear wave elastography (pSWE; 2-DSWE) using shear wave velocity (SWV) with a new machine learning (ML) technique applied to systems from different vendors is comparable to that of magnetic resonance elastography (MRE) in distinguishing non-significant (

Point Shear Wave Elastography Using Machine Learning to Differentiate Renal Cell Carcinoma and Angiomyolipoma.

The question of whether ultrasound point shear wave elastography can differentiate renal cell carcinoma (RCC) from angiomyolipoma (AML) is controversial. This study prospectively enrolled 51 patients with 52 renal tumors (42 RCCs, 10 AMLs). We obtained 10 measurements of shear wave velocity (SWV) in the renal tumor, cortex and medulla. Median SWV was first used to classify RCC versus AML. Next, the prediction accuracy of 4 machine learning algorithms-logistic regression, naïve Bayes, quadratic discriminant analysis and support vector machines (SVMs)-was evaluated, using statistical inputs from the tumor, cortex and combined statistical inputs from tumor, cortex and medulla. After leave-one-out cross validation, models were evaluated using the area under the receiver operating characteristic (ROC) curve (AUC). Tumor median SWV performed poorly (AUC = 0.62; p = 0.23). Except logistic regression, all machine learning algorithms reached statistical significance using combined statistical inputs (AUC = 0.78-0.98; p < 7.1 × 10-3). SVMs demonstrated 94% accuracy (AUC = 0.98; p = 3.13 × 10-6) and clearly outperformed median SWV in differentiating RCC from AML (p = 2.8 × 10-4).

Classification and Assessment of Hand Arthritis Stage using Support Vector Machine.

Arthritis is one of the most common health problems affecting people around the world. The goal of the work presented work is to classify and categorizing hand arthritis stages for patients, who may be developing or have developed hand arthritis, using machine learning. Stage classification was done using finger border detection, developed curvature analysis, principal components analysis, support vector machine and K-nearest neighbor algorithms. The outcome of this work showed that the proposed method can classify subject finger proximal interphalangeal joints (PIP) and distal interphalangeal joints (DIP) into stage classes with promising accuracy, especially for binary classification.

A multi-model framework to estimate perfusion parameters using contrast-enhanced ultrasound imaging.

PURPOSE: Contrast-enhanced ultrasound imaging has expanded the diagnostic potential of ultrasound by enabling real-time imaging and quantification of tissue perfusion. Several perfusion models and curve fitting methods have been developed to quantify the temporal behavior of tracer signal and standardize perfusion quantification. While the least-squares approach has traditionally been applied for curve fitting, it can be inadequate for noisy and complex data. Moreover, previous research suggests that certain perfusion models may be more relevant depending on the organ or tissue imaged. We propose a multi-model framework to select the most appropriate perfusion model and curve fitting method for each diagnostic application. METHODS: Our multi-model approach uses a system identification method, which estimates perfusion parameters from the model with the best fit to a given time-intensity curve. We compared current perfusion quantification methods that use a single perfusion model and curve fitting method and our proposed multi-model framework on bolus 3D dynamic contrast-enhanced ultrasound (DCE-US) in vivo images obtained in mice implanted with a colon cancer, as well as on simulation data. The quality of fit in estimating perfusion parameters was evaluated using the Spearman correlation coefficient, the coefficient of determination (R2 ), and the normalized root-mean-square error (NRMSE) to ensure that the multi-model framework finds the best perfusion model and curve fitting algorithm. RESULTS: Our multi-model framework outperforms conventional single perfusion model approaches with least-squares optimization, providing more robust perfusion parameter estimation. R2 and NRMSE are 0.98 and 0.18, respectively, for our proposed method. By comparison, the performance of the traditional approach is much more dependent upon the selection of the appropriate model. The R2 and NRMSE are 0.91 and 0.31, respectively. CONCLUSIONS: The proposed multi-model framework for perfusion modeling outperforms the current approach of single perfusion modeling using least-squares optimization and more robustly estimates perfusion parameters when using empiric data labeled by an expert as the gold standard. Our technique is minimally sensitive to issues affecting the accuracy of perfusion parameter estimation, including rise time, noise, region of interest size, and frame rate. This framework could be of key utility in modeling different perfusion systems in different tissues and organs.

Correlation of changes in hemodynamic response as measured by cerebral optical spectrometry with subjective pain ratings in volunteers and patients: a prospective cohort study.

PURPOSE: Noninvasive cerebral optical spectrometry is a promising candidate technology for the objective assessment physiological changes during pain perception. This study's primary objective was to test if there was a significant correlation between the changes in physiological parameters as measured by a cerebral optical spectrometry-based algorithm (real-time objective pain assessment [ROPA]) and subjective pain ratings obtained from volunteers and laboring women. Secondary aims were performance assessment using linear regression and receiver operating curve (ROC) analysis. PATIENTS AND METHODS: Prospective cohort study performed in Human Pain Laboratory and Labor and Delivery Unit. After institutional review board approval, we evaluated ROPA in volunteers undergoing the cold pressor test and in laboring women before and after epidural or combined spinal epidural placement. Linear regression was performed to measure correlations. ROCs and corresponding areas under the ROCs (AUC), as well as Youden's indices, as a measure of diagnostic effectiveness, were calculated. RESULTS: Correlations between numeric rating scale or visual analog scale and ROPA were significant for both volunteers and laboring women. AUCs for both volunteers and laboring women with numeric rating scale and visual analog scale subjective pain ratings as ground truth revealed at least good (AUC: 70%-79%) to excellent (AUC >90%) distinction between clinically meaningful pain severity differentiations (no/mild-moderate-severe). CONCLUSION: Cerebral Optical Spectrometry-based ROPA significantly correlated with subjectively reported pain in volunteers and laboring women, and could be a useful monitor for clinical circumstances where direct assessment is not available, or to complement patient-reported pain scores.

Point Shear Wave Elastography for Grading Liver Fibrosis: Can the Number of Measurements Be Reduced?

The aim of this study was to assess whether the number of liver point shear wave elastography (pSWE) measurements could be reduced compared with the currently recommended 10 valid measurements. Three thousand four hundred one pSWE examinations in patients with liver disease were performed with 10 consecutive valid measurements in liver segment 8. Liver fibrosis grading using published cutoff values were compared retrospectively using the median of 10 versus the first 1-9 measurements with Kendall's τ coefficient. Overall and binary (clinically significant [≥F2] versus non-significant [F0/F1]) fibrosis grading highly correlated when using 5-9 versus 10 valid measurements (τ = 0.96/0.95, p < 0.001). With the use of 5 valid measurements, a change in binary grading was observed in 87 of 3401 (2.6%) exams and only when velocities measured between 1.1 and 1.5 m/s. Therefore, using 5-9 valid measurements in pSWE of the liver results in a small portion of liver fibrosis grading misclassifications compared with use of 10 measurements and could help decrease scanning time, cost and discomfort in sonographers and patients.

Analyzing Seismocardiogram Cycles to Identify the Respiratory Phases.

GOAL: the objective of this study was to develop a method to identify respiratory phases (i.e., inhale or exhale) of seismocardiogram (SCG) cycles. An SCG signal is obtained by placing an accelerometer on the sternum to capture cardiac vibrations. METHODS: SCGs from 19 healthy subjects were collected, preprocessed, segmented, and labeled. To extract the most important features, each SCG cycle was divided to equal-sized bins in time and frequency domains, and the average value of each bin was defined as a feature. Support vector machines was employed for feature selection and identification. The features were selected based on the total accuracy. The identification was performed in two scenarios: leave-one-subject-out (LOSO), and subject-specific (SS). RESULTS: time-domain features resulted in better performance. The time-domain features that had higher accuracies included the characteristic points correlated with aortic-valve opening, aortic-valve closure, and the length of cardiac cycle. The average total identification accuracies were 88.1% and 95.4% for LOSO and SS scenarios, respectively. CONCLUSION: the proposed method was an efficient, reliable, and accurate approach to identify the respiratory phases of SCG cycles. SIGNIFICANCE: The results obtained from this study can be employed to enhance the extraction of clinically valuable information such as systolic time intervals.

Collagen fibers mediate MRI-detected water diffusion and anisotropy in breast cancers.

Collagen 1 (Col1) fibers play an important role in tumor interstitial macromolecular transport and cancer cell dissemination. Our goal was to understand the influence of Col1 fibers on water diffusion, and to examine the potential of using noninvasive diffusion tensor imaging (DTI) to indirectly detect Col1 fibers in breast lesions. We previously observed, in human MDA-MB-231 breast cancer xenografts engineered to fluoresce under hypoxia, relatively low amounts of Col1 fibers in fluorescent hypoxic regions. These xenograft tumors together with human breast cancer samples were used here to investigate the relationship between Col1 fibers, water diffusion and anisotropy, and hypoxia. Hypoxic low Col1 fiber containing regions showed decreased apparent diffusion coefficient (ADC) and fractional anisotropy (FA) compared to normoxic high Col1 fiber containing regions. Necrotic high Col1 fiber containing regions showed increased ADC with decreased FA values compared to normoxic viable high Col1 fiber regions that had increased ADC with increased FA values. A good agreement of ADC and FA patterns was observed between in vivo and ex vivo images. In human breast cancer specimens, ADC and FA decreased in low Col1 containing regions. Our data suggest that a decrease in ADC and FA values observed within a lesion could predict hypoxia, and a pattern of high ADC with low FA values could predict necrosis. Collectively the data identify the role of Col1 fibers in directed water movement and support expanding the evaluation of DTI parameters as surrogates for Col1 fiber patterns associated with specific tumor microenvironments as companion diagnostics and for staging.