An optimization based framework for region wise optimal clusters in MR images using hybrid objective

BackgroundIn pediatric patients who are about to undergo cochlear implantation, preimplantation imaging to assess any abnormality in the inner ear or brain is crucial. The literature has shown that inner ear malformation affects auditory and speech development. The objective of this study was to analyze the outcomes of cochlear implant (CI) in children with abnormal imaging findings.Material and methodsAll children had undergone high resolution computed tomography (HRCT) of the temporal bone and magnetic resonance imaging (MRI) of the internal acoustic canal and brain as part of routine preimplantation evaluations. Outcomes of CI were measured using Category of Auditory Performance II (CAP-II) and Meaningful Auditory Integration Scale (MAIS).ResultsThere were 25 patients who initially underwent unilateral CI, but 2 patients were excluded, leaving 23 patients in the study. There were 4 patients who had inner ear malformations and another 3 presented with brain findings, while the remaining 16 children had normal findings. In both prelingual hearing impaired children with abnormal and normal imaging findings, there were significant improvements in CAP-II score following CI (p = 0.020 and p = 0.002, respectively). Similarly, there were statistically significant in MAIS scores in prelingual hearing impaired children with abnormal imaging (p = 0.017) as well as with normal imaging (p = 0.001). However, when comparing the CAP-II and MAIS scores between children with normal and abnormal imaging, there were no significant differences observed (p = 0.624 and p = 0.376, respectively).ConclusionsThis study demonstrated that prelingually hearing impaired children with abnormal HRCT and MRI findings gained benefits from CI. Therefore, abnormal imaging findings should not be a limiting factor in CI candidacy.

Cuckoo Search (CS) algorithm was developed by Yang and Deb [1, 2] as an efficient population-based metaheuristic inspired by the behavior of some cuckoo species in combination with the Levy flight. It is also used in steel structural optimization problems by Kaveh and Bakhshpoori [3, 4]. Cuckoos are fascinating birds because of their special lifestyle and aggressive reproduction strategy. These species lay their eggs in the nests of other host birds with amazing abilities like selecting the recently spawned nests and removing existing eggs that increase the hatching probability of their eggs. The host takes care of the eggs presuming that the eggs are its own. However, some of host birds are able to combat with this parasites behavior of cuckoos and throw out the discovered alien eggs or build their new nests in new locations. Metaheuristics update the candidate solutions repetitively using the step sizes generated in the search space based on the inspired mechanisms or formulations. The randomization plays an important role in both exploration and exploitation in meta-heuristic algorithms. Therefore, these step sizes combined also with a series of consecutive random steps. In fact, the randomization part of the step sizes can vary according to a known distribution. A very special case can be achieved when the step length obeys the Levy distribution and is called Levy flight. The cuckoo breeding analogy combining with the Levy flight behavior is the essence of the CS. Each solution represents a nest or an egg or a bird. Like other metaheuristics CS starts from randomly generated initial candidate solutions. These are the eggs of host birds. In the cyclic body of the algorithm, cuckoos aim to improve the quality of the solutions by generating new eggs with step sizes toward the best known solution in combination with the Levy flight and intruded them to the nests of host birds. Moreover, in each iteration, after generating the new eggs by cuckoos, host birds have a chance to determine the alien eggs, abandon them, and generate new ones in the search space. Discovering the cuckoo’s eggs and building new nests instead of them can take place with different mechanisms. For example, it can take place for a fraction of eggs.

To develop and validate algorithms for extraction and insertion of three-dimensional (3D) profiles of tumor masses in cone beam breast CT (CBBCT) images to create simulated abnormality for evaluation with observer performance study. A bench-top experimental CBBCT scanner was constructed and used to image mastectomy breast specimens with IRB approval. 5 sets of CBBCT images with confirmed tumor masses and 2 sets of normal CBBCT images were selected and used for this study. All CBBCT images were first corrected for cupping artifacts. The corrected images were then processed to reduce their noise levels and form the denoised images. The corrected and denoised CBBCT images for normal breasts were then segmented into adipose and glandular voxels. The images for abnormal breasts were reviewed by mammographers with the help of clinical images and reports to delineate the tumor masses and form 3D tumor maps. Using these maps, the 3D tumor profiles were rescaled to average glandular signals around the insertion locations in normal images. The boundaries of the 3D rescaled tumor profiles were smoothed to avoid sharp edge. The resultant 3D tumor profiles were inserted onto cupping artifact corrected normal CBBCT images by replacing the signals in adipose voxels near the insertion locations. The cupping artifacts were then added back to generate the simulated abnormal CBBCT images. These images were then visually compared with actual abnormal images for their degree of realism. Based on visual comparison, the simulated abnormal CBBCT images showed no significant difference in realism from actual abnormal CBBCT images. We have successfully demonstrated the technique to extract 3D tumor profiles from abnormal CBBCT images and insert them onto normal CBBCT images to form simulated abnormal images for use in observer performance study. This work was supported in part by grants CA104759, CA13852 and CA124585 from the NIH-NCI, a research grant EB00117 from the NIH-NIBIB, and a subcontract from NIST-ATP.

To investigate the feasibility of training an artificial intelligence (AI) on a public-available AI platform to diagnose polypoidal choroidal vasculopathy (PCV) using indocyanine green angiography (ICGA). Two methods using AI models were trained by a data set including 430 ICGA images of normal, neovascular age-related macular degeneration (nvAMD), and PCV eyes on a public-available AI platform. The one-step method distinguished normal, nvAMD, and PCV images simultaneously. The two-step method identifies normal and abnormal ICGA images at the first step and diagnoses PCV from the abnormal ICGA images at the second step. The method with higher performance was used to compare with retinal specialists and ophthalmologic residents on the performance of diagnosing PCV. The two-step method had better performance, in which the precision was 0.911 and the recall was 0.911 at the first step, and the precision was 0.783, and the recall was 0.783 at the second step. For the test data set, the two-step method distinguished normal and abnormal images with an accuracy of 1 and diagnosed PCV with an accuracy of 0.83, which was comparable to retinal specialists and superior to ophthalmologic residents. In this evaluation of ICGA images from normal, nvAMD, and PCV eyes, the models trained on a public-available AI platform had comparable performance to retinal specialists for diagnosing PCV. The utility of public-available AI platform might help everyone including ophthalmologists who had no AI-related resources, especially those in less developed areas, for future studies.

The aim of this study was to compare the clinical characteristics of patients with and without abnormal MR imaging admitted to a neurosurgical unit with suspected cauda equina syndrome using a retrospective study of consecutive admissions to a regional neurosurgical unit over a 10-month period. Clinical details were obtained from the case notes. A lumbar spine MR scan to investigate possible cauda equina syndrome was performed in 66 patients. There were no significant differences between those with abnormal imaging (n = 34, 52%) and those with a normal scan (n = 32, 48%) in respect of sex, clinical history or features recorded on examination. Those with normal imaging had a high frequency of weakness (n = 18, 59%), saddle numbness (n = 17, 57%), leg numbness (n = 24, 80%), urinary incontinence (n = 13, 54%) and urinary retention (n = 9, 53%). A large number of patients present to neurosurgical units with symptoms suggestive of cauda equina syndrome without any radiological evidence of structural pathology. While some may have had an alternative organic cause, we propose that these symptoms may have a "functional" origin in many patients.

Event Abstract Back to Event Segmentation and analysis of hippocampus and ventricle in Alzheimer’s brain MR images using Minkowski weighted K-means clustering and its ratiometric index Kayalvizhi M1*, Kavitha G1, Sujatha C.M2 and Ramakrishnan S3 1 MIT annauniversity, Department of Electronics Engg, India 2 CEG campus Annauniversity, India 3 IIT Madras, Department of Appl. Mechanics, India Alzheimer’s disease (AD) is the common form of dementia, affecting over 24 million people worldwide with characteristic pathology of plaques and neurofibrillary tangles. This is accompanied by hippocampal atrophy and ventricular enlargement. Although AD and mild cognitive impairment (MCI) are commonly diagnosed using sets of clinical criteria, MRI findings aid the clinical diagnosis and predict clinical progression. Ventricles are located at the centre of the brain and appear dark in T1-weighted MRI scans.The hippocampus appears to be small and has regional appearance similar to its neighbouring structures. In this work, segmentation of both the hippocampus and the ventricle are carried out by Minkowski weighted K-means clustering method. The normal (30) and abnormal (30 mild, 30moderate and 30 severe) images considered in this work are obtained from MIRIAD database. Initially, the ROI is selected from original T1 sagittal image. This multiobject segmentation approach extends the K-Means criterion using Minkowski metric as a distance measure. This formulation also uses weight updating computations. Here, four initial centroids are selected for clustering from the histogram of the image.The exponent of the Minkowksi distance measure is chosen to be 6. From the clustered output, the ventricle and hippocampus are extracted using morphology operations. The segmented images are quantified using Minkowski features which captures the topology changes. These features are calculated for hippocampus and the ventricles of different normal and abnormal images and analysed.The prominent feature is correlated with the Mini-Mental State Examination (MMSE) score. Results show that Minkowski metric weighted K-Means method is able to delineate the boundary of the hippocampus and ventricle from normal and abnormal conditions.The accuracy of segmentation is high (81%). It is observed that the Minkowski area of segmented hippocampus and ventricle provide significant discrimination of normal and abnormal subjects. The ratio of hippocampus to ventricle area helps in better discrimination of severity in pathology conditions. Its correlation with MMSE is observed to be very high for normal (R=0.87) subjects.The correlation is found to be moderately high for mild (R=0.70), moderate(R=0.72) and severe (R=0.84) Alzheimer subjects. Hence this ratiometric index which takes into account of atrophy of hippocampus and the enlargement of ventricle could be used for the study of progression in neurodegenerative disorder such as AD. Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Acknowledgements Data used in the preparation of this article were obtained from the MIRIAD database (http://miriad.drc.ion.ucl.ac.uk). The MIRIAD investigators did not participate in analysis or writing of this report. The MIRIAD dataset is made available through the support of the UK Alzheimer's Society (Grant RF116). The original data collection was funded through an unrestricted educational grant from GlaxoSmithKline (Grant 6GKC).

Grey wolf optimizer (GWO) is a widely applied nature-inspired optimization algorithm with simple search mechanism, which suffers the deficiency of premature convergence. To improve its exploration performance and avoid trapping in to stagnation, we integrate the basic GWO with two random walk strategies, Gaussian random walk and Levy walk, and this new GWO extension is named RWGWO. The performance of RWGWO is compared with those of other comprising GWO variants by using several standard benchmarks. Moreover, RWGWO is applied in solving constrained engineering design problems. With a view to the experimental results, RWGWO has remarkable performance compared with the competitors in this work in terms of convergence speed and accuracy.

To evaluate the indications for transvaginal saline contrast sonohysterography (TV-SCSH) in endometrial screening by transvaginal sonography in infertile women. The study involved 850 consecutive infertile women presenting to an outpatient clinic. Using transvaginal ultrasound endometrial images were evaluated in the proliferative phase. Abnormal images were classified as follows: rugged (R), hyperechoic (H), waved (W), or thick (T). Clinical symptoms such as hypermenorrhea, dysmenorrhea and abnormal uterine bleeding were also recorded. Abnormal endometrial images were further evaluated on TV-SCSH. Age-matched women with normal endometrial images underwent TV-SCSH as controls. The endometrial pattern was abnormal in 111 patients (13.1%). Lesions that had been identified by TV-SCSH including endometrial polyps (44 cases), submucosal myomata (29 cases), and intramural myomata with mucosal extension (24 cases) were largely associated with the R and/or the H pattern, the W or the T pattern, and the W pattern, respectively. Sensitivity and specificity of the abnormal endometrial image for any lesion were 100% and 91.5%, respectively. Sixty-four patients (59.3%) were asymptomatic despite an abnormal endometrial image. TV-SCSH should be performed on selected patients following assessment of endometrial images on transvaginal sonography in order to diagnose intra- and pericavitary lesions in infertile women.

Detection of microaneurysms before diabetes increases is an essential stage in diabetic retinopathy (DR) which damages the eye, so it is big medical problem. A need arises to detect it at an early stage. It is not showing any symptoms, so it can only be a diagnosed by oculist. This paper presents the study and review of various techniques used in detection of microaneurysms as well as new approach to increasing sensitivity and reducing computational time for detection and classification of microaneurysms from the diabetic retinopathy images. This new strategy to detect MAs is based on (1) Elimination of nonuniform part of an image and standardize grayscale content of original image. (2) Performed Morphological operations for detection of Region of Interest (ROI) and elimination of blood vessels. (3) To identify real MAs two features extracted where one feature of shape which discriminates normal eye image and abnormal eye image and second feature of texture. So, this increases sensitivity and also availability. So for this whole process of new technique used publically available database called DiaretDB1 database. (4) To discriminate normal and abnormal images different clustering algorithm used.

Image segmentation is one of the pivotal steps in image processing due to its enormous application potential in medical image analysis, data mining, and pattern recognition. In fact, image segmentation is the process of splitting an image into multiple parts in order to provide detailed information on different aspects of the image. Traditional image segmentation techniques suffer from local minima and premature convergence issues when exploring complex search spaces. Additionally, these techniques also take considerable runtime to find the optimal pixels as the threshold levels are increased. Therefore, in order to overcome the computational overhead and convergence problems of the multilevel thresholding process, a robust optimizer, namely the Levy flight and Chaos theory-based Gravitational Search Algorithm (LCGSA), is employed to perform the segmentation of the COVID-19 chest CT scan images. In LCGSA, exploration is carried out by Levy flight, while chaotic maps guarantee the exploitation of the search space. Meanwhile, Kapur’s entropy method is utilized for segmenting the image into various regions based on the pixel intensity values. To investigate the segmentation performance of ten chaotic versions of LCGSA, firstly, several benchmark images from the USC-SIPI database are considered for the numerical analysis. Secondly, the applicability of LCGSA for solving real-world image processing problems is examined by using various COVID-19 chest CT scan imaging datasets from the Kaggle database. Further, an ablation study is carried out on different chest CT scan images by considering ground truth images. Moreover, various qualitative and quantitative metrics are used for the performance evaluation. The overall analysis of the experimental results indicated the efficient performance of LCGSA over other peer algorithms in terms of taking less computational time and providing optimal values for image quality metrics.

Brain tumor is generally structured by the subtle summation of anomalous cells because it categorized as a mass of tissue, and it is very crucial to categorize brain tumors from Magnetic Resonance Imaging (MRI) for diagnosis purpose. The investigation of brain tumors by humans is a routine process for detecting and classifying MRI brain tumors. Due to the low contrast of the images, poor boundaries and noises, the image classification techniques provide poor performance in medical images. Hence, a new classification model for detecting brain tumors is needed to solve the critical challenges in existing approaches. This decreases human intervention while making decisions about brain tumors. Therefore, a new brain tumor classification model is investigated to solve the challenges in existing approaches with deep learning. The proposed framework consists of two main different phases (i) classifying normal and abnormal images and (b) Segmentation and severity classification of abnormal images. The first step is comprised of pre-processing and classification. Firstly, the required images are gathered from the benchmark databases. These collected images are undergone pre-processing, which is done by filtering methods and Contrast Limited Adaptive Histogram Equalization (CLAHE). Subsequently, the pre-processed image is fed into the new classification model as a Transformer-based Multi-scale Residual Attention Network (TMRAN), where the normal and abnormal images are obtained. While in the second case, the classified abnormal image is considered as the input for the segmentation process. The segmented image is obtained by Convolutional Neural Network (CNN), Modified UNet and UNet, in turn the parameters in UNet is optimized by using the Modified Controlling Parameters-based African Vultures Optimization Algorithm (MCP-AVOA), termed as Modified UNet. At the final stage, these segmented images are applied as input to the novel TMRAN model, where the hyper parameters are tuned optimally using improved MCP-AVOA in order to acquire the optimal severity classified results. The performance is validated using diverse measures and compared over other conventional methodologies. Thus, the findings elucidate that the proposed model attains higher classification results.

Several contemporary algorithms, including cuckoo search (CS), were applied to the CEC 2017 problem set, which includes a wide variety of 120 very difficult subproblems. We found that the algorithms were ineffective, especially when the number of dimensions was high. We configured several usage patterns of Linnik flight with the inverse of the golden ratio (1/Φ) to replace Lévy flight in CS, resulting in a new search mechanism that increased the efficiency of the CS algorithm. The impacts of each Linnik flight usage pattern were evaluated using the CEC 2017. The experimental results showed that: 1) CS variants that used Linnik flight were more capable than CS variants that used Lévy flight and 2) CS variants that used a mixture of Linnik flight and quantum-behaved mechanisms were even more capable. The primary effect of Linnik flight is the strengthening of the ranking, while that of the quantum-behaved mechanisms is a decreased error. A chaotic-initialized quantum-Linnik flight CS (CQLCS) algorithm is proposed. Among the 66 competitive methods applied to the CEC 2017, CQLCS ranked first and won the contemporary competitive algorithms section, which also included several advanced variants of differential evolution (DE) and particle swarm optimization (PSO) algorithms. The CQLCS could potentially be improved further by adjusting the probability of the occurrence of Linnik flight. The processes for building improved variants were analyzed to discover how, or if, this improvement could be achieved. Finally, the CQLCS algorithm required fewer lines of code to run than did the DE and PSO variants.

We propose an algorithm for the detection of retinal landmarks (optic nerve head or optic disc, macula, and vasculature) based on optic cup location and anatomical structural details from diabetic retinopathy (DR) images of both left and right eye. Our algorithm uses color fundus images obtained from mydriatic camera. The algorithm proceeds through four main steps 1. Color image pre-processing- to enhance and remove noise from the image. 2. Detection of optic nerve head-The optic nerve head is located using a fact that the optic cup is the brightest region in the optic nerve head. At the same time exudates (DR lesion) which appear in same gray level as optic nerve head are suppressed since we only concentrate on optic cup and optic nerve head. So by calculating the mean value of the intensities of 50 ? 50 subimages (50 ? 50 is approximate area of optic cup) throughout the image and then selecting that 50 ? 50 sub image with the highest mean value, locates optic cup in the image. Using this, optic nerve head is located by increasing the area of interest around the optic cup. Since we detect optic cup first, which is embedded in optic nerve head there will not be any false detection of optic nerve head when size and shape of the exudates (DR lesion) are same as that of optic nerve head 3.Detection of macula-It is located at a distance of approximately twice the diameter of the optic nerve head just below the horizontal axis of the optic nerve head. 4. Detection of vasculature-We have used logical AND operation on two images, one being a thresholded image and another being an edge detected image. The thresholding is done on an adaptive histogram equalized image. Edge detection is done using canny edge detector. Proposed algorithm has been tested on both normal and DR images. Detected optic disc area is validated by comparing it with expert ophthalmologists' hand-drawn ground-truths. The quantitative performance is evaluated by calculating sensitivity, specificity and predictive value. Overall sensitivity (Se), specificity (Sp) and predictive value (PV) obtained in detecting optic nerve head from normal images and from abnormal images are 97.2%, 99.72%, and 88.75% and 93.93%, 99.72%, and 84.18% respectively.

The PET Imaging Site Qualification Program for amyloid positron emission tomography (PET) in Japan includes visual evaluation of the cylinder phantom. This visual evaluation requires observation of the entire image of the phantom and confirmation of the absence of apparent artifacts. Because the evaluation is visually performed, inter-observer differences might exist among evaluators for difficult cases. Therefore, the workload of the staff who perform approval tasks must be reduced, and objective evaluation methods are needed. Thus, we attempted to develop an artificial-intelligence-based objective method for anomaly detection. Three artificial intelligence methods for anomaly detection were developed, and their accuracy was evaluated using AutoEncoder, AnoGAN, and a method combining feature extraction using AlexNet and a one-class support vector machine. In total, 10,207 normal images from 128 facilities and 594 abnormal images from eight facilities, all of which were submitted as part of application for amyloid PET certification, were used. Group five-fold cross-validation was employed for artificial intelligence training and evaluation. In addition, the performance of each artificial intelligence method was assessed using receiver operating characteristic analysis. The areas under the curve for anomaly detection using AutoEncoder, AnoGAN, and the method combining feature extraction using AlexNet and a one-class support vector machine were 0.80 ± 0.04, 0.77 ± 0.03, and 0.99 ± 0.01, respectively. Artificial intelligence effectively distinguished between normal and abnormal images with high accuracy. In the future, its practical implementation is anticipated to reduce the workload in the approval work for the Japanese site qualification program for amyloid PET.

Background: Erdheim Chester disease (ECD) is a rare histiocytic neoplasm associated with hypothalamic and pituitary infiltration and dysfunction. We determined the abnormal pituitary imaging (API) phenotypes in subjects with ECD and analyzed their associated endocrine dysfunctions. Methods: This was a cross-sectional examination of a natural history cohort study of 61 subjects with ECD performed at a tertiary care clinical research center. The diagnosis of ECD was based on clinical, molecular, and histopathological features. Enrolled subjects underwent baseline endocrine tests of anterior and posterior pituitary function in addition to pituitary imaging. The following variables were analyzed- age, sex, body mass index (BMI), BRAF V600E, hsCRP, ESR, pituitary hormone deficit number, diabetes insipidus (DI), and panhypopituitarism. Fisher’s exact test or t-test/Wilcoxon tests compared patients with and without API. Results: Sixty-one subjects with ECD (age ±SD: 54.3 ±10.9, 46 (75.4%) males) were studied. The prevalence of API was 32.8% (n=20), who were younger than those with normal imaging (50.3 ±10.5 vs 56.3 ±10.7 yrs, p=0.042). The most common pituitary imaging abnormalities included thickened pituitary stalk (18.03%, n=11/61), followed by pituitary encasement, small pituitary and abnormal morphology (6.55%, n= 4/61 for each). A higher prevalence of DI (45.0% vs 9.8%, p=0.003) and panhypopituitarism (45.0% vs 4.9%, p<0.001), and a higher number of pituitary deficits (median (IQR): 2.0 (0-2.5) vs 0 (0-1.0), p=0.007) were noted in patients with API compared to those with normal imaging. Other biochemical markers were similar between both groups. Conclusion: Abnormal pituitary imaging was commonly seen in ECD and was also associated with a younger age and hormone deficits suggesting associations with pituitary structure-function.