Prediction of Near-Term Breast Cancer Occurrence using Subtraction of Temporally Sequential Digital Mammograms.

Breast cancer remains one of the leading cancers for women worldwide. Fortunately, with the introduction of mammography, the mortality rate has significantly decreased. However, earlier breast cancer prediction could effectively increase the survival rates, improve patient outcomes, and avoid unnecessary biopsies. For that purpose, prediction of breast cancer, using subtraction of temporally sequential digital mammograms and machine learning, is proposed. A new dataset was collected with 192 images from 32 patients (three screening rounds, with two views of each breast). This dataset included precise annotation of each individual malignant mass, present in the most recent mammogram, with the two priors being radiologically evaluated as normal. The most recent mammogram was considered as the "future" screening round and provided the location of the mass as the ground truth for the training. The two previous mammograms, the "current" and the "prior", were processed and a new, difference image was formed for the prediction. Ninety-six features were extracted and five feature selection algorithms were combined to identify the most important features. Ten classifiers were tested in leave-one-patient-out and k-fold-patient cross-validation (k = 4 and 8). Ensemble Voting achieved the highest performance in the prediction of the development of breast mass in the next screening round, with 85.7% sensitivity, 83.7% specificity, 83.7% accuracy and 0.85 AUC. The proposed methodology could lead to a new mammography-based model that could predict the short-term risk for developing a malignancy, thus providing an earlier diagnosis.

Computer-aided breast cancer detection and classification in mammography: A comprehensive review.

Cancer is the second cause of mortality worldwide and it has been identified as a perilous disease. Breast cancer accounts for ∼20% of all new cancer cases worldwide, making it a major cause of morbidity and mortality. Mammography is an effective screening tool for the early detection and management of breast cancer. However, the identification and interpretation of breast lesions is challenging even for expert radiologists. For that reason, several Computer-Aided Diagnosis (CAD) systems are being developed to assist radiologists to accurately detect and/or classify breast cancer. This review examines the recent literature on the automatic detection and/or classification of breast cancer in mammograms, using both conventional feature-based machine learning and deep learning algorithms. The review begins with a comparison of algorithms developed specifically for the detection and/or classification of two types of breast abnormalities, micro-calcifications and masses, followed by the use of sequential mammograms for improving the performance of the algorithms. The available Food and Drug Administration (FDA) approved CAD systems related to triage and diagnosis of breast cancer in mammograms are subsequently presented. Finally, a description of the open access mammography datasets is provided and the potential opportunities for future work in this field are highlighted. The comprehensive review provided here can serve both as a thorough introduction to the field but also provide indicative directions to guide future applications.

Automatic Breast Mass Segmentation and Classification Using Subtraction of Temporally Sequential Digital Mammograms.

OBJECTIVE: Cancer remains a major cause of morbidity and mortality globally, with 1 in 5 of all new cancers arising in the breast. The introduction of mammography for the radiological diagnosis of breast abnormalities, significantly decreased their mortality rates. Accurate detection and classification of breast masses in mammograms is especially challenging for various reasons, including low contrast and the normal variations of breast tissue density. Various Computer-Aided Diagnosis (CAD) systems are being developed to assist radiologists with the accurate classification of breast abnormalities. METHODS: In this study, subtraction of temporally sequential digital mammograms and machine learning are proposed for the automatic segmentation and classification of masses. The performance of the algorithm was evaluated on a dataset created especially for the purposes of this study, with 320 images from 80 patients (two time points and two views of each breast) with precisely annotated mass locations by two radiologists. RESULTS: Ninety-six features were extracted and ten classifiers were tested in a leave-one-patient-out and k-fold cross-validation process. Using Neural Networks, the detection of masses was 99.9% accurate. The classification accuracy of the masses as benign or suspicious increased from 92.6%, using the state-of-the-art temporal analysis, to 98%, using the proposed methodology. The improvement was statistically significant (p-value < 0.05). CONCLUSION: These results demonstrate the effectiveness of the subtraction of temporally consecutive mammograms for the diagnosis of breast masses. Clinical and Translational Impact Statement: The proposed algorithm has the potential to substantially contribute to the development of automated breast cancer Computer-Aided Diagnosis systems with significant impact on patient prognosis.

A Review of Computer-Aided Breast Cancer Diagnosis Using Sequential Mammograms.

Radiologists assess the results of mammography, the key screening tool for the detection of breast cancer, to determine the presence of malignancy. They, routinely, compare recent and prior mammographic views to identify changes between the screenings. In case a new lesion appears in a mammogram, or a region is changing rapidly, it is more likely to be suspicious, compared to a lesion that remains unchanged and it is usually benign. However, visual evaluation of mammograms is challenging even for expert radiologists. For this reason, various Computer-Aided Diagnosis (CAD) algorithms are being developed to assist in the diagnosis of abnormal breast findings using mammograms. Most of the current CAD systems do so using only the most recent mammogram. This paper provides a review of the development of methods to emulate the radiological approach and perform automatic segmentation and/or classification of breast abnormalities using sequential mammogram pairs. It begins with demonstrating the importance of utilizing prior views in mammography, through the review of studies where the performance of expert and less-trained radiologists was compared. Following, image registration techniques and their application to mammography are presented. Subsequently, studies that implemented temporal analysis or subtraction of temporally sequential mammograms are summarized. Finally, a description of the open access mammography datasets is provided. This comprehensive review can serve as a thorough introduction to the use of prior information in breast cancer CAD systems but also provides indicative directions to guide future applications.

Identification and Classification of Benign and Malignant Masses based on Subtraction of Temporally Sequential Digital Mammograms.

Breast cancer remains the leading cause of cancer deaths and the second highest cause of death, in general, among women worldwide. Fortunately, over the last few decades, with the introduction of mammography, the mortality rate of breast cancer has significantly decreased. However, accurate classification of breast masses in mammograms is especially challenging. Various Computer-Aided Diagnosis (CAD) systems are being developed to assist radiologists with the accurate classification of breast abnormalities. In this study, classification of benign and malignant masses, based on the subtraction of temporally sequential digital mammograms and machine learning, is proposed. The performance of the algorithm was evaluated on a dataset created for the purposes of this study. In total, 196 images from 49 patients, with precisely annotated mass locations and biopsy confirmed malignant cases, were included. Ninety-six features were extracted and five feature selection algorithms were employed to identify the most important features. Ten classifiers were tested using leave-one-patient-out and 7-fold cross-validation. Neural Networks, achieved the highest classification performance with 90.85% accuracy and 0.91 AUC, an improvement compared to the state-of-the-art. These results demonstrate the effectiveness of the subtraction of temporally consecutive mammograms for the classification of breast masses as benign or malignant.

Digital subtraction of temporally sequential mammograms for improved detection and classification of microcalcifications.

BACKGROUND: Our aim was to demonstrate that automated detection and classification of breast microcalcifications, according to Breast Imaging Reporting and Data System (BI-RADS) categorisation, can be improved with the subtraction of sequential mammograms as opposed to using the most recent image only. METHODS: One hundred pairs of mammograms were retrospectively collected from two temporally sequential rounds. Fifty percent of the images included no (BI-RADS 1) or benign (BI-RADS 2) microcalcifications. The remaining exhibited suspicious findings (BI-RADS 4-5) in the recent image. Mammograms cannot be directly subtracted, due to tissue changes over time and breast deformation during mammography. To overcome this challenge, optimised preprocessing, image registration, and postprocessing procedures were developed. Machine learning techniques were employed to eliminate false positives (normal tissue misclassified as microcalcifications) and to classify the true microcalcifications as BI-RADS benign or suspicious. Ninety-six features were extracted and nine classifiers were evaluated with and without temporal subtraction. The performance was assessed by measuring sensitivity, specificity, accuracy, and area under the curve (AUC) at receiver operator characteristics analysis. RESULTS: Using temporal subtraction, the contrast ratio improved ~ 57 times compared to the most recent mammograms, enhancing the detection of the radiologic changes. Classifying as BI-RADS benign versus suspicious microcalcifications, resulted in 90.3% accuracy and 0.87 AUC, compared to 82.7% and 0.81 using just the most recent mammogram (p = 0.003). CONCLUSION: Compared to using the most recent mammogram alone, temporal subtraction is more effective in the microcalcifications detection and classification and may play a role in automated diagnosis systems.

Dual-angle optical coherence tomography for index of refraction estimation using rigid registration and cross-correlation.

The index of refraction (n) of materials and/or tissues depends on their physical properties and serves as a source of optical contrast in imaging. The variations of the index of refraction have also been investigated for diagnostic purposes in various fields, such as hematology, oncology, etc., since they can signify disease and cell dynamic changes. Optical coherence tomography (OCT) has been used in the past to measure the index ex vivo. However, most methodologies described in the literature are not appropriate for in vivo imaging since they require either a mirror below the sample or a complicated imaging setup and algorithms. We describe a technique that uses two images, obtained at different angles, to estimate the index of refraction and can, thus, also be applied in vivo. The index of refraction is calculated from the path-length difference observed by the OCT beam at the two different angles. When a reflector is not available, the path-length difference can be estimated using image registration and the cross-correlation of adjacent A-scans. The proposed technique was validated experimentally using both clear and scattering samples. The resulting values of the index of refraction were within ∼1 % of the expected. The main limitation of this technique is the effect of misalignment on the results, requiring the precision provided by an angular-resolved OCT system. These very promising results provide evidence that the dual-angle method should be further investigated and validated on human tissues so that it can be developed into a clinically useful diagnostic tool in the future.

A Novel Conjugate of Bis[((4-bromophenyl)amino)quinazoline], a EGFR-TK Ligand, with a Fluorescent Ru(II)-Bipyridine Complex Exhibits Specific Subcellular Localization in Mitochondria.

The epidermal growth factor receptor (EGFR) is a key target in anticancer research, whose aberrant function in malignancies has been linked to severe irregularities in critical cellular processes, including cell cycle progression, proliferation, differentiation, and survival. EGFR mutant variants, either transmembrane or translocated to the mitochondria and/or the nucleus, often exhibit resistance to EGFR inhibitors. The ability to noninvasively image and quantify EGFR provides novel approaches in the detection, monitoring, and treatment of EGFR-related malignancies. The current study aimed to deliver a new theranostic agent that combines fluorescence imaging properties with EGFR inhibition. This was achieved via conjugation of an in-house-developed ((4-bromophenyl)amino)quinazoline inhibitor of mutant EGFR-TK, selected from a focused aminoquinazoline library, with a [Ru(bipyridine)3]2+ fluorophore. A triethyleneglycol-derived diamino linker featuring (+)-ionizable sites was employed to link the two functional moieties, affording two unprecedented Ru conjugates with 1:1 and 2:1 stoichiometry of aminoquinazoline to the Ru complex (mono-quinazoline-Ru-conjugate and bis-quinazoline-Ru-conjugate, respectively). The bis-quinazoline-Ru-conjugate, which retains an essential inhibitory activity, was found by fluorescence imaging to be effectively uptaken by Uppsala 87 malignant glioma (grade IV malignant glioma) cells. The fluorescence imaging study and a time-resolved fluorescence resonance energy transfer study indicated a specific subcellular distribution of the conjugate that coincides with that of a mitochondria-targeted dye, suggesting mitochondrial localization of the conjugate and potential association with mitochondria-translocated forms of EGFR. Mitochondrial localization was further documented by the specific concentration of the bis-quinazoline-Ru-conjugate in a mitochondrial isolation assay.

Comparison of tissue dispersion measurement techniques based on optical coherence tomography.

The effects of dispersion on optical coherence tomography (OCT) images have long been documented. The imbalance of spectral broadening, caused by dispersion mismatches in the two arms of the OCT interferometer, can result in significant resolution degradation. Efforts to correct this phenomenon have resulted in improved image quality using various techniques. However, dispersion is also present and varies in tissues. As a result, group velocity dispersion (GVD) can be used to detect changes in tissues and provide useful information for diagnosis. Several methods can be utilized to measure the GVD from OCT images: (i) the degradation of the point spread function (PSF), (ii) the shift (walk-off) between images taken at different wavelengths, (iii) the changes in the second derivative of the spectral phase, as well as two new methods, which do not require a reflector and are applicable in intact tissues, i.e., using (iv) the speckle degradation, and (v) the speckle cross correlation. A systematic, experimental, evaluation of these methods is presented to elucidate the capabilities, the limitations, and the accuracy of each technique when attempting to estimate the GVD in scattering samples. The most precise values were obtained from the estimation of the PSF degradation, whereas using the phase derivative method was only applicable to minimally scattering samples. Speckle broadening appears to be the most robust method for tissue GVD measurements.

Radiogenomics for Precision Medicine With a Big Data Analytics Perspective.

Precision medicine promises better healthcare delivery by improving clinical practice. Using evidence-based substratification of patients, the objective is to achieve better prognosis, diagnosis, and treatment that will transform existing clinical pathways toward optimizing care for the specific needs of each patient. The wealth of today's healthcare data, often characterized as big data, provides invaluable resources toward new knowledge discovery that has the potential to advance precision medicine. The latter requires interdisciplinary efforts that will capitalize the information, know-how, and medical data of newly formed groups fusing different backgrounds and expertise. The objective of this paper is to provide insights with respect to the state-of-the-art research in precision medicine. More specifically, our goal is to highlight the fundamental challenges in emerging fields of radiomics and radiogenomics by reviewing the case studies of Cancer and Alzheimer's disease, describe the computational challenges from a big data analytics perspective, and discuss standardization and open data initiatives that will facilitate the adoption of precision medicine methods and practices.

Lateral resolution improvement of oversampled OCT images using Capon estimation of weighted subvolume contribution.

A novel technique for lateral resolution improvement in optical coherence tomography (OCT) is presented. The proposed method is based on lateral oversampling of the image. The locations and weights of multiple high spatial resolution sub-volumes are calculated using a Capon estimator assuming each contributes a weighted portion to the detected signal. This technique is independent of the delivery optics and the depth of field. Experimental results demonstrate that it is possible to achieve ~4x lateral resolution improvement which can be diagnostically valuable, especially in cases where the delivery optics are constrained to low numerical aperture (NA).

Using speckle to measure tissue dispersion in optical coherence tomography.

Tissue dispersion could be used as a marker of early disease changes to further improve the diagnostic potential of optical coherence tomography (OCT). However, most methods to measure dispersion, described in the literature, rely on the presence of distinct and strong reflectors and are, therefore, rarely applicable in vivo. A novel technique has been developed which estimates the dispersion-induced resolution degradation from the image speckle and, as such, is applicable in situ. This method was verified experimentally ex vivo and was applied to the classification of a set of normal and cancerous colon OCT images resulting in 96% correct classification.

Axial resolution improvement by modulated deconvolution in Fourier domain optical coherence tomography.

A novel technique for axial resolution improvement in Fourier domain optical coherence tomography (FDOCT) is presented. The technique is based on the deconvolution of modulated optical coherence tomography signals. In FDOCT, the real part of the Fourier transform of the interferogram is modulated by a frequency which depends on the position of the interferogram in k space. A slight numerical k shift results in a different modulation frequency. By adding two shifted signals, beating can appear in the A-scan. When the amount of shifting is appropriately selected, deconvolution of the resulting depth profile, using suitable modulated kernels, yields a narrower resolution width. A resolution improvement by a factor of ∼7 can be achieved without the need for a broader bandwidth light source.

Scatterer size-based analysis of optical coherence tomography images using spectral estimation techniques.

A novel spectral analysis technique of OCT images is demonstrated in this paper for classification and scatterer size estimation. It is based on SOCT autoregressive spectral estimation techniques and statistical analysis. Two different statistical analysis methods were applied to OCT images acquired from tissue phantoms, the first method required prior information on the sample for variance analysis of the spectral content. The second method used k-means clustering without prior information for the sample. The results are very encouraging and indicate that the spectral content of OCT signals can be used to estimate scatterer size and to classify dissimilar areas in phantoms and tissues with sensitivity and specificity of more than 90%.

Optical coherence tomography axial resolution improvement by step-frequency encoding.

A novel technique for axial resolution improvement of Optical Coherence Tomography (OCT) systems is proposed. The technique is based on step-frequency encoding, using frequency shifting, of the OCT signal. A resolution improvement by a factor of approximately 7 is achieved without the need for a broader bandwidth light source. This method exploits a combination of two basic principles: the appearance of beating, when adding two signals of slightly different carrier frequencies, and the resolution improvement by deconvolution of the interferogram with an encoded autocorrelation function. In time domain OCT, step-frequency encoding can be implemented by performing two scans, with different carrier frequencies, and subsequently adding them to create the encoded signal. When the frequency steps are properly selected, deconvolution of the resulting interferogram, using appropriate kernels, results in a narrower resolution width.

AM-FM techniques in the analysis of optical coherence tomography signals.

The subtle tissue changes associated with the early stages of malignancies, such as cancer, are not clearly discernible even at the current, improved, resolution of optical coherence tomography (OCT) systems. However, these changes directly affect the spectral content of the OCT image that contains information regarding these unresolvable features. Spectral analysis of OCT signals has recently been shown to provide additional information, resulting in improved contrast, directly related to scatterer size changes. Amplitude modulation-frequency modulation (AM-FM) analysis, a fast and accurate technique for the estimation of the instantaneous frequency, phase, and amplitude of a signal, can also be applied to OCT images to extract scatterer-size information. The proposed technique could make available an extremely valuable tool for the investigation of disease characteristics that now remain below the resolution of OCT and could significantly improve the technology's diagnostic capabilities.

Transillumination spatially modulated illumination microscopy.

The diagnostic utility of a conventional transillumination microscope, the most common imaging modality in clinical use today, is limited by the microscope's resolution. It is, however, possible to achieve lateral resolution well beyond the classical limit by using laterally structured illumination in a wide-field, nonconfocal microscope. In this method, the spatially modulated illumination (SMI) makes high-resolution information that is normally inaccessible visible in the observed image. Previously presented SMI microscopy systems operated in epifluorescence mode. We describe the design, construction, and testing of a novel transillumination SMI microscope. As transillumination is necessary for most medical applications, such as histopathologic evaluation of biopsy tissue and chromosomal analysis, such a system should have a significant diagnostic effect.

High-resolution optical coherence tomographic imaging of osteoarthritic cartilage during open knee surgery.

This study demonstrates the first real-time imaging in vivo of human cartilage in normal and osteoarthritic knee joints at a resolution of micrometers, using optical coherence tomography (OCT). This recently developed high-resolution imaging technology is analogous to B-mode ultrasound except that it uses infrared light rather than sound. Real-time imaging with 11-microm resolution at four frames per second was performed on six patients using a portable OCT system with a handheld imaging probe during open knee surgery. Tissue registration was achieved by marking sites before imaging, and then histologic processing was performed. Structural changes including cartilage thinning, fissures, and fibrillations were observed at a resolution substantially higher than is achieved with any current clinical imaging technology. The structural features detected with OCT were evident in the corresponding histology. In addition to changes in architectural morphology, changes in the birefringent or the polarization properties of the articular cartilage were observed with OCT, suggesting collagen disorganization, an early indicator of osteoarthritis. Furthermore, this study supports the hypothesis that polarization-sensitive OCT may allow osteoarthritis to be diagnosed before cartilage thinning. This study illustrates that OCT, which can eventually be developed for use in offices or through an arthroscope, has considerable potential for assessing early osteoarthritic cartilage and monitoring therapeutic effects for cartilage repair with resolution in real time on a scale of micrometers.