Automated analysis of low-field brain MRI in cerebral malaria

A central challenge of medical imaging studies is to extract biomarkers that characterize disease pathology or outcomes. Modern automated approaches have found tremendous success in high-resolution, high-quality magnetic resonance images. These methods, however, may not translate to low-resolution images acquired on magnetic resonance imaging (MRI) scanners with lower magnetic field strength. In low-resource settings where low-field scanners are more common and there is a shortage of radiologists to manually interpret MRI scans, it is critical to develop automated methods that can augment or replace manual interpretation, while accommodating reduced image quality. We present a fully automated framework for translating radiological diagnostic criteria into image-based biomarkers, inspired by a project in which children with cerebral malaria (CM) were imaged using low-field 0.35 Tesla MRI. We integrate multiatlas label fusion, which leverages high-resolution images from another sample as prior spatial information, with parametric Gaussian hidden Markov models based on image intensities, to create a robust method for determining ventricular cerebrospinal fluid volume. We also propose normalized image intensity and texture measurements to determine the loss of gray-to-white matter tissue differentiation and sulcal effacement. These integrated biomarkers have excellent classification performance for determining severe brain swelling due to CM.     

A Review of the Deep Learning Methods for Medical Images Super Resolution Problems

Super resolution problems are widely discussed in medical imaging. Spatial resolution of medical images are not sufficient due to the constraints such as image acquisition time, low irradiation dose or hardware limits. To address these problems, different super resolution methods have been proposed, such as optimization or learning-based approaches. Recently, deep learning methods become a thriving technology and are developing at an exponential speed. We think it is necessary to write a review to present the current situation of deep learning in medical imaging super resolution. In this paper, we first briefly introduce deep learning methods, then present a number of important deep learning approaches to solve super resolution problems, different architectures as well as up-sampling operations will be introduced. Afterwards, we focus on the applications of deep learning methods in medical imaging super resolution problems, the challenges to overcome will be presented as well.     

Pattern recognition software and techniques for biological image analysis

The increasing prevalence of automated image acquisition systems is enabling new types of microscopy experiments that generate large image datasets. However, there is a perceived lack of robust image analysis systems required to process these diverse datasets. Most automated image analysis systems are tailored for specific types of microscopy, contrast methods, probes, and even cell types. This imposes significant constraints on experimental design, limiting their application to the narrow set of imaging methods for which they were designed. One of the approaches to address these limitations is pattern recognition, which was originally developed for remote sensing, and is increasingly being applied to the biology domain. This approach relies on training a computer to recognize patterns in images rather than developing algorithms or tuning parameters for specific image processing tasks. The generality of this approach promises to enable data mining in extensive image repositories, and provide objective and quantitative imaging assays for routine use. Here, we provide a brief overview of the technologies behind pattern recognition and its use in computer vision for biological and biomedical imaging. We list available software tools that can be used by biologists and suggest practical experimental considerations to make the best use of pattern recognition techniques for imaging assays.     

An Objective Method to Optimize the MR Sequence Set for Plaque Classification in Carotid Vessel Wall Images Using Automated Image Segmentation

A typical MR imaging protocol to study the status of atherosclerosis in the carotid artery consists of the application of multiple MR sequences. Since scanner time is limited, a balance has to be reached between the duration of the applied MR protocol and the quantity and quality of the resulting images which are needed to assess the disease. In this study an objective method to optimize the MR sequence set for classification of soft plaque in vessel wall images of the carotid artery using automated image segmentation was developed. The automated method employs statistical pattern recognition techniques and was developed based on an extensive set of MR contrast weightings and corresponding manual segmentations of the vessel wall and soft plaque components, which were validated by histological sections. Evaluation of the results from nine contrast weightings showed the tradeoff between scan duration and automated image segmentation performance. For our dataset the best segmentation performance was achieved by selecting five contrast weightings. Similar performance was achieved with a set of three contrast weightings, which resulted in a reduction of scan time by more than 60%. The presented approach can help others to optimize MR imaging protocols by investigating the tradeoff between scan duration and automated image segmentation performance possibly leading to shorter scanning times and better image interpretation. This approach can potentially also be applied to other research fields focusing on different diseases and anatomical regions.     

Automated classification of healthy and keloidal collagen patterns based on processing of SHG images of human skin

All‐optical microspectroscopic and tomographic tools have a great potential for the clinical investigation of human skin and skin diseases. However, automated optical tomography or even microscopy generate immense data sets. Therefore, in order to implement such diagnostic tools into the medical practice in both hospitals and private practice, there is a need for automated data handling and image analysis ideally implementing automized scores to judge the physiological state of a tissue section. In this contribution, the potential of an image processing algorithm for the automated classification of skin into normal or keloid based on second‐harmonic generation (SHG) microscopic images is demonstrated. Such SHG data is routinely recorded within a multimodal imaging approach. The classification of the tissue implemented in the algorithm employs the geometrical features of collagen patterns that differ depending on the constitution, i.e., physiological status of the skin. (© 2011 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

Bridging the imaging gap: visualizing subcellular architecture with electron tomography

Transmission electron microscopy is a powerful tool that is used to explore the internal structure of tissues, cells, organelles and macromolecular complexes. By integrating data from a series of images in which the orientation of the specimen is progressively varied relative to the incident electron beam it is also possible to extend electron microscopic imaging into the third dimension. This approach, commonly referred to as electron tomography, has been greatly aided in recent years by advances in technology for imaging specimens at cryogenic temperatures, as well as by substantial progress in procedures for automated data collection and image processing. The intense pace of developments in this field is inspired, in a large part, by the hope that the quality of the data will ultimately be good enough to allow interpretation of tomograms of cells, organelles, bacteria and viruses in terms of the three-dimensional spatial arrangements of the constituent molecules.     

Habitat-Net: Segmentation of habitat images using deep learning

Understanding environmental factors that influence forest health, as well as the occurrence and abundance of wildlife, is a central topic in forestry and ecology. However, the manual processing of field habitat data is time-consuming and months are often needed to progress from data collection to data interpretation. To shorten the time to process the data we propose here Habitat-Net: a novel deep learning application based on Convolutional Neural Networks (CNN) to segment habitat images of tropical rainforests. Habitat-Net takes color images as input and after multiple layers of convolution and deconvolution, produces a binary segmentation of the input image. We worked on two different types of habitat datasets that are widely used in ecological studies to characterize the forest conditions: canopy closure and understory vegetation. We trained the model with 800 canopy images and 700 understory images separately and then used 149 canopy and 172 understory images to test the performance of Habitat-Net. We compared the performance of Habitat-Net to the performance of a simple threshold based method, manual processing by a second researcher and a CNN approach called U-Net, upon which Habitat-Net is based. Habitat-Net, U-Net and simple thresholding reduced total processing time to milliseconds per image, compared to 45 s per image for manual processing. However, the higher mean Dice coefficient of Habitat-Net (0.94 for canopy and 0.95 for understory) indicates that accuracy of Habitat-Net is higher than that of both the simple thresholding (0.64, 0.83) and U-Net (0.89, 0.94). Habitat-Net will be of great relevance for ecologists and foresters, who need to monitor changes in their forest structures. The automated workflow not only reduces the time, it also standardizes the analytical pipeline and, thus, reduces the degree of uncertainty that would be introduced by manual processing of images by different people (either over time or between study sites).     

Auto-align - multi-modality fluorescence microscopy image co-registration

Multi-modality microscopes incorporate multiple microscopy techniques into one module, imaging through a common objective lens. Simultaneous or consecutive image acquisition of a single specimen, using multiple techniques, increases the amount of measurable information available. In order to benefit from each modality, it is necessary to accurately co-register data sets. Intrinsic differences in the image formation process employed by each modality result in images which possess different characteristics. In addition, as a result of using different measurement devices, images often differ in size and can suffer relative geometrical deformations including rotation, scale and translation, making registration a complex problem. Current methods generally rely on manual input and are therefore subject to human error. Here, we present an automated image registration tool for fluorescence microscopy. We show that it successfully registers images obtained via total internal reflection fluorescence (TIRF), or epi-fluorescence, and confocal microscopy. Furthermore, we provide several other applications including channel merging following image acquisition through an emission beam splitter, and lateral stage drift correction. We also discuss areas of membrane trafficking which could benefit from application of Auto-Align. Auto-Align is an essential item in the advanced microscopist's toolbox which can create a synergy of single or multi-modality image data.     

A Microfluidic Platform for Correlative Live-Cell and Super-Resolution Microscopy

Recently, super-resolution microscopy methods such as stochastic optical reconstruction microscopy (STORM) have enabled visualization of subcellular structures below the optical resolution limit. Due to the poor temporal resolution, however, these methods have mostly been used to image fixed cells or dynamic processes that evolve on slow time-scales. In particular, fast dynamic processes and their relationship to the underlying ultrastructure or nanoscale protein organization cannot be discerned. To overcome this limitation, we have recently developed a correlative and sequential imaging method that combines live-cell and super-resolution microscopy. This approach adds dynamic background to ultrastructural images providing a new dimension to the interpretation of super-resolution data. However, currently, it suffers from the need to carry out tedious steps of sample preparation manually. To alleviate this problem, we implemented a simple and versatile microfluidic platform that streamlines the sample preparation steps in between live-cell and super-resolution imaging. The platform is based on a microfluidic chip with parallel, miniaturized imaging chambers and an automated fluid-injection device, which delivers a precise amount of a specified reagent to the selected imaging chamber at a specific time within the experiment. We demonstrate that this system can be used for live-cell imaging, automated fixation, and immunostaining of adherent mammalian cells in situ followed by STORM imaging. We further demonstrate an application by correlating mitochondrial dynamics, morphology, and nanoscale mitochondrial protein distribution in live and super-resolution images.     

Medical ultrasound imaging

The paper gives an introduction to current medical ultrasound imaging systems. The basics of anatomic and blood flow imaging are described. The properties of medical ultrasound and its focusing are described, and the various methods for two- and three-dimensional imaging of the human anatomy are shown. Systems using both linear and non-linear propagation of ultrasound are described. The blood velocity can also be non-invasively visualized using ultrasound and the basic signal processing for doing this is introduced. Examples for spectral velocity estimation, color flow imaging and the new vector velocity images are presented.     

Artificial intelligence and machine learning for medical imaging: A technology review

Artificial intelligence (AI) has recently become a very popular buzzword, as a consequence of disruptive technical advances and impressive experimental results, notably in the field of image analysis and processing. In medicine, specialties where images are central, like radiology, pathology or oncology, have seized the opportunity and considerable efforts in research and development have been deployed to transfer the potential of AI to clinical applications. With AI becoming a more mainstream tool for typical medical imaging analysis tasks, such as diagnosis, segmentation, or classification, the key for a safe and efficient use of clinical AI applications relies, in part, on informed practitioners. The aim of this review is to present the basic technological pillars of AI, together with the state-of-the-art machine learning methods and their application to medical imaging. In addition, we discuss the new trends and future research directions. This will help the reader to understand how AI methods are now becoming an ubiquitous tool in any medical image analysis workflow and pave the way for the clinical implementation of AI-based solutions.     

Recent enhancements in SSCP

High sensitivity, robustness and scalability are the three criteria which influence whether techniques for rapid mutation detection will be used in the future. PCR-SSCP, one of the most popular methods for detecting mutation, especially in the field of medical genetics, is being improved (1) to efficiently detect mutations in long stretches of PCR products; (2) to simplify data interpretation by removing PCR artifacts and (3) to minimize human involvement in the process of mutation detection by a simple post-PCR fluorescence labeling followed by separation using automated DNA sequencers.     

Computer aided diagnosis of Coronary Artery Disease,Myocardial Infarction and carotid atherosclerosis using ultrasound images: A review

The diagnosis of Coronary Artery Disease (CAD), Myocardial Infarction (MI) and carotid atherosclerosis is of paramount importance, as these cardiovascular diseases may cause medical complications and large number of death. Ultrasound (US) is a widely used imaging modality, as it captures moving images and image features correlate well with results obtained from other imaging methods. Furthermore, US does not use ionizing radiation and it is economical when compared to other imaging modalities. However, reading US images takes time and the relationship between image and tissue composition is complex. Therefore, the diagnostic accuracy depends on both time taken to read the images and experience of the screening practitioner. Computer support tools can reduce the inter-operator variability with lower subject specific expertise, when appropriate processing methods are used. In the current review, we analysed automatic detection methods for the diagnosis of CAD, MI and carotid atherosclerosis based on thoracic and Intravascular Ultrasound (IVUS). We found that IVUS is more often used than thoracic US for CAD. But for MI and carotid atherosclerosis IVUS is still in the experimental stage. Furthermore, thoracic US is more often used than IVUS for computer aided diagnosis systems.     

Pigeons (Columba livia) as Trainable Observers of Pathology and Radiology Breast Cancer Images

Pathologists and radiologists spend years acquiring and refining their medically essential visual skills, so it is of considerable interest to understand how this process actually unfolds and what image features and properties are critical for accurate diagnostic performance. Key insights into human behavioral tasks can often be obtained by using appropriate animal models. We report here that pigeons (Columba livia)—which share many visual system properties with humans—can serve as promising surrogate observers of medical images, a capability not previously documented. The birds proved to have a remarkable ability to distinguish benign from malignant human breast histopathology after training with differential food reinforcement; even more importantly, the pigeons were able to generalize what they had learned when confronted with novel image sets. The birds’ histological accuracy, like that of humans, was modestly affected by the presence or absence of color as well as by degrees of image compression, but these impacts could be ameliorated with further training. Turning to radiology, the birds proved to be similarly capable of detecting cancer-relevant microcalcifications on mammogram images. However, when given a different (and for humans quite difficult) task—namely, classification of suspicious mammographic densities (masses)—the pigeons proved to be capable only of image memorization and were unable to successfully generalize when shown novel examples. The birds’ successes and difficulties suggest that pigeons are well-suited to help us better understand human medical image perception, and may also prove useful in performance assessment and development of medical imaging hardware, image processing, and image analysis tools.     

Human vs. machine: Detecting wildlife in camera trap images

As the capacity to collect and store large amounts of data expands, identifying and evaluating strategies to efficiently convert raw data into meaningful information is increasingly necessary. Across disciplines, this data processing task has become a significant challenge, delaying progress and actionable insights. In ecology, the growing use of camera traps (i.e., remotely triggered cameras) to collect information on wildlife has led to an enormous volume of raw data (i.e., images) in need of review and annotation. To expedite camera trap image processing, many have turned to the field of artificial intelligence (AI) and use machine learning models to automate tasks such as detecting and classifying wildlife in images. To contribute understanding of the utility of AI tools for processing wildlife camera trap images, we evaluated the performance of a state-of-the-art computer vision model developed by Microsoft AI for Earth named MegaDetector using data from an ongoing camera trap study in Arctic Alaska, USA. Compared to image labels determined by manual human review, we found MegaDetector reliably determined the presence or absence of wildlife in images generated by motion detection camera settings (≥94.6% accuracy), however, performance was substantially poorer for images collected with time-lapse camera settings (≤61.6% accuracy). By examining time-lapse images where MegaDetector failed to detect wildlife, we gained practical insights into animal size and distance detection limits and discuss how those may impact the performance of MegaDetector in other systems. We anticipate our findings will stimulate critical thinking about the tradeoffs of using automated AI tools or manual human review to process camera trap images and help to inform effective implementation of study designs.     

La géométrie fractale pour l’analyse de signaux médicaux : état de l’art

The diagnostic interpretation of medical images is a complex task aiming to detect potential abnormalities. One of the most used features in this process is texture which is a key component in the human understanding of images. Many studies were conducted to develop algorithms for texture quantification. The relevance of fractal geometry in medical image analysis is justified by the proven self-similarity of anatomical objects when imaged with a finite resolution. Over the last years, fractal geometry was applied extensively in many medical signal analysis applications. The use of these geometries relies heavily on estimation of the fractal features. Various methods were proposed to estimate the fractal dimension or multifractal spectrum of a signal. This article presents an overview of these algorithms, the way they work, their benefits and limits, and their application in the field of medical signal analysis.     

Image analysis for monitoring the barley tempeh fermentation process

AIMS: To develop a fast, accurate, objective and nondestructive method for monitoring barley tempeh fermentation. METHODS AND RESULTS: Barley tempeh is a food made from pearled barley grains fermented with Rhizopus oligosporus. Rhizopus oligosporus growth is important for tempeh quality, but quantifying its growth is difficult and laborious. A system was developed for analysing digital images of fermentation stages using two image processing methods. The first employed statistical measures sensitive to image colour and surface structure, and these statistical measures were highly correlated (r=0.92, n=75, P<0.001) with ergosterol content of tempeh fermented with R. oligosporus and lactic acid bacteria (LAB). In the second method, an image-processing algorithm optimized to changes in images of final tempeh products was developed to measure number of visible barley grains. A threshold of 5 visible grains per Petri dish indicated complete tempeh fermentation. When images of tempeh cakes fermented with different inoculation levels of R. oligosporus were analysed the results from the two image processing methods were in good agreement. CONCLUSION: Image processing proved suitable for monitoring barley tempeh fermentation. The method avoids sampling, is nonintrusive, and only requires a digital camera with good resolution and image analysis software. SIGNIFICANCE AND IMPACT OF THE STUDY: The system provides a rapid visualization of tempeh product maturation and qualities during fermentation. Automated online monitoring of tempeh fermentation by coupling automated image acquisition with image processing software could be further developed for process control.     

An Adaptive Low-Rank Modeling-Based Active Learning Method for Medical Image Annotation

Active learning is an effective solution to interactively select a limited number of informative examples and use them to train a learning algorithm that can achieve its optimal performance for specific tasks. It is suitable for medical image applications in which unlabeled data are abundant but manual annotation could be very time-consuming and expensive. However, designing an effective active learning strategy for informative example selection is a challenging task, due to the intrinsic presence of noise in medical images, the large number of images, and the variety of imaging modalities. In this study, a novel low-rank modeling-based multi-label active learning (LRMMAL) method is developed to address these challenges and select informative examples for training a classifier to achieve the optimal performance. The proposed method independently quantifies image noise and integrates it with other measures to guide a pool-based sampling process to determine the most informative examples for training a classifier. In addition, an automatic adaptive cross entropy-based parameter determination scheme is proposed for further optimizing the example sampling strategy. Experimental results on varied medical image datasets and comparisons with other state-of-the-art multi-label active learning methods illustrate the superior performance of the proposed method.     

Determination of three-dimensional imaging properties of a light microscope system. Partial confocal behavior in epifluorescence microscopy.

We have determined the three-dimensional image-forming properties of an epifluorescence microscope for use in obtaining very high resolution three-dimensional images of biological structures by image processing methods. Three-dimensional microscopic data is collected as a series of two-dimensional images recorded at different focal planes. Each of these images contains not only in-focus information from the region around the focal plane, but also out-of-focus contributions from the remainder of the specimen. Once the imaging properties of the microscope system are characterized, powerful image processing methods can be utilized to remove the out-of-focus information and to correct for image distortions. Although theoretical calculations for the behavior of an aberration-free microscope system are available, the properties of real lenses under the conditions used for biological observation are often far from an ideal. For this reason, we have directly determined the image-forming properties of an epifluorescence microscope under conditions relevant to biological observations. Through-focus series of a point object (fluorescently-coated microspheres) were recorded on a charge-coupled device image detector. From these images, the three-dimensional point spread function and its Fourier transform, the optical transfer function, were derived. There were significant differences between the experimental results and the theoretical models which have important implications for image processing. The discrepancies can be explained by imperfections of the microscope system, nonideal observation conditions, and partial confocal effects found to occur with epifluorescence illumination. Understanding the optical behavior of the microscope system has indicated how to optimize specimen preparation, data collection, and processing protocols to obtain significantly improved images.