Morphological Image Processing Techniques Research Articles

Abstract 4815: Humanized single mouse trial: A preclinical platform feasible for immune-oncology drug screening and translational biomarker development

Abstract The field of cancer immunology is rapidly moving towards innovative therapeutic strategies. As a consequence the need for robust and predictive preclinical platforms arises just as well. The current project aims to establish a drug screening workflow bridging between innovative mouse models and clinical biomarker development. A total of 69 NOG (NOD/Shi-scid/IL-2Rγnull) mice were engrafted with CD34+ hematopoietic stem cells. Thereafter, tumor material from 11 different lung cancer patient derived xenograft models (NSCLC PDX) was implanted subcutaneously. Individual mice were treated with α-CTLA-4, α-PD-1 or the combination thereof. With n=1 per treatment arm and model the study design followed the screening approach of the single mouse trial (SMT). Infiltration of human immune cells was detected by flow cytometry (FC) and immunohistochemistry (IHC) in hematopoietic organs and tumor tissue. A computerized analysis for digitized whole-slide images of the samples was used to quantify the lymphocyte infiltration using color classification and morphological image processing techniques. All 3 treatment arms displayed a discrete activity pattern throughout the PDX panel. Tumor models with high tumor infiltrating lymphocyte (TIL) rates in the donor patient material tended to be more sensitive towards checkpoint inhibitor treatment as models with low rates. Numbers of TILs in the PDX detected by FC and IHC were significantly increased in the treatment groups as compared to control vehicle. In parallel, hematopoietic organs showed high (>25%) amounts of huCD45 cells in all groups and models. PDX models being sensitive towards checkpoint inhibitor treatment (responders) displayed a higher percentage of DAB+ nuclei in huCD45 IHC stains than non-responder models as determined by image analysis. Irrespective thereof, in responders as well as non-responders the treatment with checkpoint inhibitors enhanced the percentage of DAB+ nuclei. Whole-slide image analysis of the H&E stains revealed an increase of the stromal compartment proportion in the tumor tissue under treatment with checkpoint inhibitors in responder models. In non-responder models the ratio between tumor and stroma was not influenced by drug treatment. The use of PDX based humanized mouse models in a SMT format allows screening approaches in complex mouse models. The combination with a comprehensive image analysis tool enables additional read-outs to quantify antitumoral activity of immune modulatory compounds. The latter can be used to identify possible biomarkers in the preclinical setting. Moreover, the translation and validation of these biomarker candidates in a clinical setting is self-evident as primary material needed for these types of analyses is easily accessible. Citation Format: Daniel Bug, Eva Oswald, Anne Grote, Anne-Lise Peille, Gabriele Niedermann, Dorit Merhof, Friedrich Feuerhake, Julia B. Schüler. Humanized single mouse trial: A preclinical platform feasible for immune-oncology drug screening and translational biomarker development [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 4815. doi:10.1158/1538-7445.AM2017-4815

Machine vision for counting fruit on mango tree canopies

Machine vision technologies hold the promise of enabling rapid and accurate fruit crop yield predictions in the field. The key to fulfilling this promise is accurate segmentation and detection of fruit in images of tree canopies. This paper proposes two new methods for automated counting of fruit in images of mango tree canopies, one using texture-based dense segmentation and one using shape-based fruit detection, and compares the use of these methods relative to existing techniques:—(i) a method based on K-nearest neighbour pixel classification and contour segmentation, and (ii) a method based on super-pixel over-segmentation and classification using support vector machines. The robustness of each algorithm was tested on multiple sets of images of mango trees acquired over a period of 3 years. These image sets were acquired under varying conditions (light and exposure), distance to the tree, average number of fruit on the tree, orchard and season. For images collected under the same conditions as the calibration images, estimated fruit numbers were within 16 % of actual fruit numbers, and the F1 measure of detection performance was above 0.68 for these methods. Results were poorer when models were used for estimating fruit numbers in trees of different canopy shape and when different imaging conditions were used. For fruit-background segmentation, K-nearest neighbour pixel classification based on colour and smoothness or pixel classification based on super-pixel over-segmentation, clustering of dense scale invariant feature transform features into visual words and bag-of-visual-word super-pixel classification using support vector machines was more effective than simple contrast and colour based segmentation. Pixel classification was best followed by fruit detection using an elliptical shape model or blob detection using colour filtering and morphological image processing techniques. Method results were also compared using precision–recall plots. Imaging at night under artificial illumination with careful attention to maintaining constant illumination conditions is highly recommended.

Identification of Diabetic Maculopathy Stages using Fundus Images

The impairment and blindness are caused due to Diabetes mellitus. Most of sick person are suffering with impairment diabetes and fifty to sixty percent of ill persons are suffering due to blindness diabetes. Present paper concentrated the identification process of diabetes mellitus using a computer-based intelligent system. The method consists of non-clinically and clinically significant maculopathy and usual eye images. With the help of morphological image processing techniques the fundus are extracted from the usual eye images. Later for comparison the extracted data is applied to classifiers named as (i) feed-forward ANN and (ii) (PNN) probabilistic neural network.

Automatic crack detection and classification method for subway tunnel safety monitoring.

Cracks are an important indicator reflecting the safety status of infrastructures. This paper presents an automatic crack detection and classification methodology for subway tunnel safety monitoring. With the application of high-speed complementary metal-oxide-semiconductor (CMOS) industrial cameras, the tunnel surface can be captured and stored in digital images. In a next step, the local dark regions with potential crack defects are segmented from the original gray-scale images by utilizing morphological image processing techniques and thresholding operations. In the feature extraction process, we present a distance histogram based shape descriptor that effectively describes the spatial shape difference between cracks and other irrelevant objects. Along with other features, the classification results successfully remove over 90% misidentified objects. Also, compared with the original gray-scale images, over 90% of the crack length is preserved in the last output binary images. The proposed approach was tested on the safety monitoring for Beijing Subway Line 1. The experimental results revealed the rules of parameter settings and also proved that the proposed approach is effective and efficient for automatic crack detection and classification.

Automated detection of lung cancer using statistical and morphological image processing techniques

Lung cancer represents the second most commonly diagnosed cancer among Jordanian population. Evidence that early detection of lung cancer may allow for more timely therapeutic intervention has provided the momentum for lung cancer screening programs around the world. In this study, a computer aided detection (CAD) system is proposed in an attempt to detect the lung cancer areas using computed tomography (CT) images. It is implemented as a “second reader” to help radiologists focus their attention on regions that might be missed during visual interpretation. The proposed CAD system has three main stages; Segmentation by thresholding the CT images, labeling the founded regions and then extracting some diagnostic features of each region for further analysis and interpretation. The study is trained, tested, and validated using images obtained from forty five patients. The obtained results perfectly match the radiologist's diagnosis in detecting the defected areas and quantitatively measuring its size, location, borders as well as displaying its other diagnostic characteristics. Moreover, the proposed system can detect misclassified regions.

Digitization of Spörer's sunspot drawings

AbstractMuch of our knowledge about the solar dynamo is based on sunspot observations. It is thus desirable to extend the set of positional and morphological data of sunspots into the past. Gustav Spörer observed in Germany from Anklam (1861–1873) and Potsdam (1874–1894). He left detailed prints of sunspot groups, which we digitized and processed to mitigate artifacts left in the print by the passage of time. After careful geometrical correction, the sunspot data are now available as synoptic charts for almost 450 solar rotation periods. Individual sunspot positions can thus be precisely determined and spot areas can be accurately measured using morphological image processing techniques. These methods also allow us to determine tilt angles of active regions (Joy's law) and to assess the complexity of an active region.

Morphological analysis of vibrational hyperspectral imaging data

This study demonstrates the use of standard morphological image processing techniques to reduce the hyperspectral image data of samples, containing discrete particles or domains, to a single average spectrum per particle. The processing is automated and successful even when the particles are in contact. Focal Plane Array, Fourier transform infrared (FTIR) absorbance images of biological cells are used as an example dataset. The large number of spectra in the image (~40,000) can be intelligently averaged to ~100 mean spectra, approximately one per cell, greatly simplifying further analysis. As well as reducing the data, the morphological analysis provides useful information, such as the size of each cell, and allows every spectrum associated with each cell to be identified and analysed independently of the full dataset. Using these methods, combined with principal components analysis, consistent spectral differences are found between the spectra of the whole cells and a cell region approximately corresponding to the nucleus. These spectral differences compare well with previous IR measurements on whole CALU-1 cells and their isolated nuclei, but with a simpler sample preparation. The algorithm created to analyse the CALU-1 cells has been applied to a second cell line (NL20), which has a very different growth morphology, to demonstrate that this processing method is applicable to varied samples with little or no modification.

Monitoring changes in the Menik Farm IDP camps in Sri Lanka using multi-temporal very high-resolution satellite data

This paper presents the results of the application of mathematical morphology for the automatic quantification of the number of tents and the detection of changes in the Menik Farm IDP camp in Sri Lanka. The tents were retrieved using an area-constraint top-hat opening applied to WorldView-1 data. The counting of tents was based on the centroids of the automatically detected structures indicating an overall number of 25,150 tents on the 26 June 2009. The comparison with a visual interpretation produced an R 2 of 0.97 with an error of 1.25%. In addition, an automated detection of changes inside a camp area was conducted. The comparison of the satellite image of 26 June 2009 (WorldView-1) and an image of 28 February 2010 (GeoEye-1) is based on mutual (mixed) information metric, after using morphological image processing techniques and previously specified criterion. Changes are observed on a terrain of around 15.2% of the total camp area and 3813 of previously detected structures disappeared in a period of 8 months.

Investigation of polynya dynamics in the northern Bering Sea using greyscale morphology image-processing techniques

Arctic coastal polynyas, defined here as persistent openings in the winter sea ice pack, are significant areas for ocean–atmosphere heat exchange, high winter sea ice production with resulting brine rejection and the dependence of local ecosystems. It is therefore critical to accurately quantify polynya dynamics to understand their spatial and temporal variability, particularly in the context of recent dramatic Arctic sea ice declines. In this study, the so-called erosion algorithm (a greyscale morphology image-processing technique) is applied to satellite-derived sea ice concentrations in the northern Bering Sea to investigate polynya dynamics throughout this region. Greyscale morphology allows for the estimation of sea ice extent with a defined error tolerance through the removal of regions with low sea ice concentrations in the marginal ice zone. Furthermore, since polynyas are the primary source of water within the sea ice pack, the presence of water here can therefore be utilized to define the areal extent of polynyas. We utilized AMSR-E (Advanced Microwave Scanning Radiometer-EOS) sea ice concentrations during January–April from 2003 to 2008 in the northern Bering Sea to extract daily time series of water area within polynyas by a water integration method after data are preprocessed using greyscale morphology techniques. These results compare well with those calculated by more traditional methods utilizing sea ice concentration thresholds and show the great utility of greyscale morphology techniques as a preprocessing method (which eliminates artificial determination of polynya areal extents and enables automation of the overall image-processing routine). In addition, based on the results of our algorithms we investigated the potential driving forces (e.g. offshore wind velocity) of polynya development in the northern Bering Sea as well as calculate the spatial and interannual variability of heat fluxes across these water surfaces owing to polynya formation.

A quick moving target detection method based on real-time airborne videos

In this paper, a fast detection method of moving target from real-time airborne videos is presented. We first explore the motion relationship in an airborne video and utilise the global motion estimation to estimate the background movement, and then use binocular interpolation for motion compensation; thus, the moving target detection in dynamic airborne videos is converted into detection in static backgrounds. A method involving the multiplication of the differences of the symmetrical images among three adjacent frames is adopted to enhance moving target and increase the difference between the target and background noises. The regional growth and morphological image processing techniques are employed finally to realise target segmentation. Experimental results demonstrate the effectiveness and robustness of our algorithm.

Automatic Detection of Limb Prominences in 304 Å EUV Images

A new algorithm for automatic detection of prominences on the solar limb in 304 Å EUV images is presented, and results of its application to SOHO/EIT data discussed. The detection is based on the method of moments combined with a classifier analysis aimed at discriminating between limb prominences, active regions, and the quiet corona. This classifier analysis is based on a Support Vector Machine (SVM). Using a set of 12 moments of the radial intensity profiles, the algorithm performs well in discriminating between the above three categories of limb structures, with a misclassification rate of 7%. Pixels detected as belonging to a prominence are then used as the starting point to reconstruct the whole prominence by morphological image-processing techniques. It is planned that a catalogue of limb prominences identified in SOHO and STEREO data using this method will be made publicly available to the scientific community.

Experimental Study of Particle Motion on a Smooth Bed under Shoaling Waves Using Particle Image Velocimetry

The motion of spherical particles (diameter 1.58 mm, specific gravity 2.5) on 2 and 3% plane slope was studied in a laboratory wave flume for shoaling wave conditions. The range of wave-height-to-water-depth ratio was 0.24<H/h<0.82 , the range of Keulegan-Carpenter number was 128<KC<500 , and the maximum particle Reynolds number was 240. The instantaneous sediment particle position and the associated fluid velocity field were measured simultaneously in the bottom boundary layer using a particle image velocimetry (PIV) system. The measurement plane was parallel to the bed and located at a height of 1/2 particle diameter above the bed. Morphological image processing techniques were used to separate the fluid (tracers) and solid (glass spheres) phases from the same PIV images based on their signature sizes. The separated image files were processed to determine the displacement, velocity, and acceleration of the individual sediment particles and the corresponding fluid velocity and total acceleration. These m...

Computer-based detection of diabetes retinopathy stages using digital fundus images

Diabetes mellitus is a heterogeneous clinical syndrome characterized by hyperglycaemia and the long-term complications are retinopathy, neuropathy, nephropathy, and cardiomyopathy. It is a leading cause of blindness. Diabetic retinopathy is the progressive pathological alterations in the retinal microvasculature, leading to areas of retinal nonperfusion, increased vascular permeability, and the pathological proliferation of retinal vessels. Hence, it is beneficial to have regular cost-effective eye screening for diabetes subjects. Nowadays, different stages of diabetes retinopathy are detected by retinal examination using indirect biomicroscopy by senior ophthalmologists. In this work, morphological image processing and support vector machine (SVM) techniques were used for the automatic diagnosis of eye health. In this study, 331 fundus images were analysed. Five groups were identified: normal retina, mild non-proliferative diabetic retinopathy, moderate non-proliferative diabetic retinopathy, severe non-proliferative diabetic retinopathy, and proliferative diabetic retinopathy. Four salient features blood vessels, microaneurysms, exudates, and haemorrhages were extracted from the raw images using image-processing techniques and fed to the SVM for classification. A sensitivity of more than 82 per cent and specificity of 86 per cent was demonstrated for the system developed.

Automatic Segmentation of the Ribs, the Vertebral Column, and the Spinal Canal in Pediatric Computed Tomographic Images

We propose methods to perform automatic identification of the rib structure, the vertebral column, and the spinal canal in computed tomographic (CT) images of pediatric patients. The segmentation processes for the rib structure and the vertebral column are initiated using multilevel thresholding and the results are refined using morphological image processing techniques with features based on radiological and anatomical prior knowledge. The Hough transform for the detection of circles is applied to a cropped edge map that includes the thoracic vertebral structure. The centers of the detected circles are used to derive the information required for the opening-by-reconstruction algorithm used to segment the spinal canal. The methods were tested on 39 CT exams of 13 patients; the results of segmentation of the vertebral column and the spinal canal were assessed quantitatively and qualitatively by comparing with segmentation performed independently by a radiologist. Using 13 CT exams of six patients, including a total of 458 slices with the vertebra from different sections of the vertebral column, the average Hausdorff distance was determined to be 3.2 mm with a standard deviation (SD) of 2.4 mm; the average mean distance to the closest point (MDCP) was 0.7 mm with SD = 0.6 mm. Quantitative analysis was also performed for the segmented spinal canal with three CT exams of three patients, including 21 slices with the spinal canal from different sections of the vertebral column; the average Hausdorff distance was 1.6 mm with SD = 0.5 mm, and the average MDCP was 0.6 mm with SD = 0.1 mm.

Robust estimation of tear film surface quality in lateral shearing interferometry

Interferometry is a sensitive technique for recording tear film surface irregularities in a noninvasive manner. At the same time, the technique is hindered by natural eye movements resulting in measurement noise. Estimating tear film surface quality from interferograms can be reduced to a spatial-average-localized weighted estimate of the first harmonic of the interference fringes. However, previously reported estimation techniques proved to perform poorly in cases where the pattern fringes were significantly disturbed. This can occur in cases of measuring tear film surface quality on a contact lens on the eye or in a dry eye. We present a new estimation technique for extracting the first harmonic from the interference fringes that combines the traditional spectral estimation techniques with morphological image processing techniques. The proposed technique proves to be more robust to changes in interference fringes caused by natural eye movements and the degree of dryness of the contact lens and corneal surfaces than its predecessors, resulting in tear film surface quality estimates that are less noisy.

A Preliminary Investigation of Avian Comb as a Potential Biometric Marker for Identification of Poultry

In today's globalized world, to ensure effective traceability, food-producing animals must be identified by a tamper-proof and durable technique. This article investigates the avian comb profile as a potential biometric-based method for the traceability of poultry. Forty mature hens were used for this experiment. Video was acquired of comb profiles, and morphological image processing techniques including dilation and erosion were performed on still images obtained from the video with the purpose of linking a segmented template or fingerprint with each bird that was imaged to act as a unique identifier. Comb profile features were abstracted using the Zahn-Roskies Fourier descriptor technique for subsequent shape analysis. Multivariate discriminant analysis was applied to the most discriminant harmonic variables, resulting in a positive classification rate of 80.6%. When a simple comb overlap function was combined with the Fourier technique, the rate increased to 84.4%.

Robust approach for suburban road segmentation in high‐resolution aerial images

The goal of this research is to develop an algorithm that accurately segments high‐resolution images, where linear features, such as roads, are corrupted by noise. In high‐resolution images, there are two types of noises that are obstacles to road segmentation. Noise could be within road areas (such as cars, water, differences in composite road surface mix) or unwanted contents outside road areas, like buildings and trees. To remove unwanted contents, the geographical information from the United States Geographical Survey (USGS) is used. The USGS provides a collection of road centre line information that has been collected for many years and can be used to limit the area for road segmentations close to roads. In this paper, a standard process was developed to align the USGS geographical information with the high‐resolution images. USGS geographical data is used to eliminate background clutter that is disjunct from roads. The road segmentation process is then reduced to dealing with automobile traffic, shadows and pavement colour discontinuity within road areas. In order to achieve reliable road segmentation in the presence of these objects, the mean‐shift clustering approach is used within the hue‐saturation‐intensity (HSI) space. Conditional morphological image processing techniques are also used to significantly improve the segmentation results. The proposed method results in the average accuracy of road segmentation above 85%.

Morphological phase unwrapping

Phase unwrapping is the process of determining the absolute phase given its principal value. In this paper, we show that morphological image processing techniques can be helpful for solving this problem. The proposed procedure consists of three independent modules. The first aims at extracting a region of interest corresponding to the areas where interferences occur. This is achieved by processing the input interferograms with a combination of morphological opening and closing operators. The second finds the high boundaries of the wrapped fringes by enhancing them with a morphological gradient by erosion. Gaps along the fringe boundaries which would lead to errors in the unwrapped phase map are filled in by a morphological region growing segmentation technique. The third module consists in integrating the wrapped phase while taking into account the segmented fringes.

Use of morphological image processing techniques for the measurement of a fetal head from ultrasound images

An automatic method is described which uses a class of nonlinear image processing filters known as morphological operators to connect and isolate the fetal head within an ultrasound image and generate a measurement of its diameter. This measurement is a valuable indicator in estimating the gestational age of the infant. The advantage of this approach is that it provides both an automatic and a consistent method to determine these measurements. Currently the measurement is made by hand and is highly subjective. A comparison with the existing manual approach is presented.

Lip-motion analysis for speech segmentation in noise

This paper explains how visual information from the lips and acoustic signals can be combined together for speech segmentation. The psychological aspects of lip-reading and current automatic lip-reading systems are reviewed. The paper describes an image processing system which can extract the velocity of the lips from image sequences. The velocity of the lips is estimated by a combination of morphological image processing and block matching techniques. The resultant velocity of the lips is used to locate the syllable boundaries. This information is particularly useful when the speech signal is corrupted by noise. The paper also demonstrates the correlation between speech signals and lip information. Data fusion techniques are used to combine the acoustic and visual information for speech segmentation. The principal results show that using the combination of visual and acoustic signals can reduce segmentation errors by at least 10.4% when the signal-to-noise ratio is lower than 15 dB.