Published research results are difficult to replicate due to the lack of a standard evaluation data set in the area of decision support systems in mammography; most computer-aided diagnosis (CADx) and detection (CADe) algorithms for breast cancer in mammography are evaluated on private data sets or on unspecified subsets of public databases. This causes an inability to directly compare the performance of methods or to replicate prior results. We seek to resolve this substantial challenge by releasing an updated and standardized version of the Digital Database for Screening Mammography (DDSM) for evaluation of future CADx and CADe systems (sometimes referred to generally as CAD) research in mammography. Our data set, the CBIS-DDSM (Curated Breast Imaging Subset of DDSM), includes decompressed images, data selection and curation by trained mammographers, updated mass segmentation and bounding boxes, and pathologic diagnosis for training data, formatted similarly to modern computer vision data sets. The data set contains 753 calcification cases and 891 mass cases, providing a data-set size capable of analyzing decision support systems in mammography.

/pdf/a-curated-mammography-data-set-for-use-in-computer-aided-2g8dyfxrop.pdf

A curated mammography data set for use in computer-aided detection and diagnosis research.

Frontotemporal dementia (FTD) and Alzheimer disease (AD) can be difficult to clinically differentiate because of overlapping symptoms.1–3 Structural MRI can assess regional patterns of brain pathology and may aid a differential diagnosis between FTD and AD.4–10 However, overlapping of brain tissue loss limits structural MRI to achieve complete separation between FTD and AD. Functional neuroimaging, yielding complementary information to tissue loss, may help the differentiation between FTD and AD. Previous PET and SPECT studies have shown different patterns of metabolism and perfusion reduction in FTD and AD.11–26 However, PET and SPECT have several disadvantages, including injection of radioactive tracers and limited availability in comparison with MRI scanners. Arterial spin labeled perfusion MRI (ASL-MRI) is another method to assess brain perfusion.27 Previous studies found that ASL-MRI detects a regional pattern of hypoperfusion in AD,28,29 similar to PET and SPECT. Therefore, we tested if ASL-MRI detects a regional pattern of hypoperfusion in FTD vs cognitively normal (CN) subjects and patients with AD, similar to PET and SPECT.

A complication in interpreting functional imaging data is that brain atrophy and partial volume effects (PVEs) can mimic metabolic and perfusion reductions. Therefore, we also sought to determine if the patterns of hypoperfusion in FTD remain after accounting for PVE. Last, we explored if ASL-MRI can depict the severity of FTD and if PVE-corrected perfusion can improve the separation of FTD from CN and AD over that using structural MRI alone.

/pdf/hypoperfusion-in-frontotemporal-dementia-and-alzheimer-413jw5swhq.pdf

Hypoperfusion in frontotemporal dementia and Alzheimer disease by arterial spin labeling MRI.

Angiogenesis has long been established as a key element in the pathophysiology of tumor growth and metastasis. Increasingly, new molecularly targeted antiangiogenic drugs are being developed in the fight against cancer. These drugs bring with them a need for an accurate means of diagnosing tumor angiogenesis and monitoring response to treatment. Imaging techniques can offer this in a noninvasive way, while also providing functional information about the tumor. Among the many clinical imaging techniques available, MRI alone can provide relatively good spatial resolution and specificity, without ionizing radiation and with limited side effects. Arterial spin labeling (ASL) and blood oxygenation level-dependent (BOLD) imaging techniques can be employed to confer indirect measures of angiogenesis, such as blood flow and blood volume, without the need for external contrast agents. Dynamic contrast-enhanced (DCE)-MRI is rapidly emerging as a standard method for directly measuring angiogenesis during angiogenesis-inhibitor drug trials. As macromolecular MR contrast agents become available, they will inevitably be utilized in the assessment of tumor perfusion and vessel permeability. Meanwhile, technological advances have made imaging at a molecular level a possibility. They have brought the potential to directly target MR contrast agents to markers of angiogenesis, such as the alpha(v)beta(3) integrin. Before this is used clinically, however, substantial gains in sensitivity brought about by improved coils, pulse sequences, and contrast agents will be needed. Herein we discuss the techniques currently available for MRI of angiogenesis, along with their respective advantages and disadvantages, and what the future holds for this evolving field of imaging.

https://onlinelibrary.wiley.com/doi/pdf/10.1002/jmri.20991

MRI of tumor angiogenesis.

Arterial spin labeling (ASL) offers MRI measurement of cerebral blood flow (CBF) in vivo, and may offer clinical diagnostic utility in populations such as those with early Alzheimer's disease (AD). In the current study, we investigated the reliability and precision of a pseudo-continuous ASL (pcASL) sequence that was performed two or three times within one hour on eight young normal control subjects, and 14 elderly subjects including 11 with normal cognition, one with AD and two with Mild Cognitive Impairment (MCI). Six of these elderly subjects including one AD, two MCIs and three controls also received (15)O-water positron emission tomography (PET) scans 2 h before their pcASL MR scan. The instrumental reliability of pcASL was evaluated with the intraclass correlation coefficient (ICC). The ICCs were greater than 0.90 in pcASL global perfusion measurements for both the young and the elderly groups. The cross-modality perfusion imaging comparison yielded very good global and regional agreement in global gray matter and the posterior cingulate cortex. Significant negative correlation was found between age and the gray/white matter perfusion ratio (r = -0.62, p < 0.002). The AD and MCI patients showed the lowest gray/white matter perfusion ratio among all the subjects. The data suggest that pcASL provides a reliable whole brain CBF measurement in young and elderly adults whose results converge with those obtained with the traditional (15)O-water PET perfusion imaging method. pcASL perfusion MRI offers an alternative method for non-invasive in vivo examination of early pathophysiological changes in AD.

/pdf/reliability-and-precision-of-pseudo-continuous-arterial-spin-2cdaaa1ixq.pdf

Reliability and Precision of Pseudo-Continuous Arterial Spin Labeling Perfusion MRI on 3.0 T and Comparison With 15O-water PET in Elderly Subjects at Risk for Alzheimer's Disease

A deep learning approach for the analysis of masses in mammograms with minimal user intervention.

The Committee of Human Research of the University of California San Francisco approved this study, and all volunteers provided written informed consent. The goal of this study was to prospectively determine the global and regional reliability and reproducibility of noninvasive brain perfusion measurements obtained with different pulsed arterial spin-labeling (ASL) magnetic resonance (MR) imaging methods and to determine the extent to which within-subject variability and random noise limit reliability and reproducibility. Thirteen healthy volunteers were examined twice within 2 hours. The pulsed ASL methods compared in this study differ mainly with regard to magnetization transfer and eddy current effects. There were two main results: (a) Pulsed ASL MR imaging consistently had high measurement reliability (intraclass correlation coefficients greater than 0.75) and reproducibility (coefficients of variation less than 8.5%), and (b) random noise rather than within-subject variability limited reliability and reproducibility. It was concluded that low signal-to-noise ratios substantially limit the reliability and reproducibility of perfusion measurements.

/pdf/human-brain-reliability-and-reproducibility-of-pulsed-3sombov5ce.pdf

Human brain: reliability and reproducibility of pulsed arterial spin-labeling perfusion MR imaging.

Denoising 3D MR images by the enhanced non-local means filter for Rician noise

Breast mass segmentation in mammography using plane fitting and dynamic programming.

B-scan ultrasound is the primary means for the diagnosis of fatty liver. However, due to use of various ultrasound equipments, poor quality of ultrasonic images and physical differences of patients, fatty liver diagnosis is mainly qualitative, and often depends on the subjective judgment of technicians and doctors. Therefore, computer-aided feature extraction and quantitative analysis of liver B-scan ultrasonic images will help to improve clinical diagnostic accuracy, repeatability and efficiency, and could provide a measure for severity of hepatic steatosis. This paper proposed a novel method of fatty liver diagnosis based on liver B-mode ultrasonic images using support vector machine (SVM). Fatty liver diagnosis was transformed into a pattern recognition problem of liver ultrasound image features. According to the different characteristics of fatty liver and healthy liver, important image features were extracted and selected to distinguish between the two categories. These features could be represented by near-field light-spot density, near-far-field grayscale ratio, grayscale co-occurrence matrix, and neighborhood gray-tone difference matrix (NGTDM). A SVM classifier was modeled and trained using the clinical ultrasound images of both fatty liver and normal liver. It was then exploited to classify normal and fatty livers, achieving a high recognition rate. The diagnostic results are satisfactorily consistent with those made by doctors. This method could be used for computer-aided diagnosis of fatty liver, and help doctors identify the fatty liver ultrasonic images rapidly, objectively and accurately.

Computer aided diagnosis of fatty liver ultrasonic images based on support vector machine

Background/purpose: Skin microrelief has been studied using various methods and devices. However, the long duration of time needed to process one sample or the expensive equipment hampered the use of those systems for routine diagnosis. Today, the emergence of new software and hardware technologies may allow this issue to be resolved.



Methods: To characterize objectively the skin surface, we introduced a new parameter SPm, namely, the area mean of superficial skin texture block formed by primary and secondary lines crossing each other. Based on the skin detector produced by the Boseview Technology Company, we developed a software for acquiring automatically skin images and calculating SPm. The relationship between SPm and age was studied on the dorsal and ventral midway of the forearm (sun-exposed and sun-protected areas) of 94 healthy volunteers without a history of smoking.



Results: The skin surface topography can be conveniently quantified with the new parameter SPm. The value of SPm of both sites increases with age, independent of sex, with the site more exposed to light being more affected.



Conclusion: With the software developed, the details of the skin surface can be observed. SPm appears to be a new valid parameter for characterizing the property of the skin surface. Our method, alone or in combination with other technologies of skin topography analysis may be applied in routine diagnosis for a quantified evaluation of skin aging.

Enmin Song

Papers

Human brain: reliability and reproducibility of pulsed arterial spin-labeling perfusion MR imaging.

Denoising 3D MR images by the enhanced non-local means filter for Rician noise

Breast mass segmentation in mammography using plane fitting and dynamic programming.

Computer aided diagnosis of fatty liver ultrasonic images based on support vector machine

Age-dependent changes in skin surface assessed by a novel two-dimensional image analysis.