Ahmed M. Ghanem

Three different classifiers for facial age estimation based on K-nearest neighbor

The exact age estimation is often treated as a classification problem; while it can be formulated as a regression problem. In this article, three different classifiers based on KNN classifiers concept for facial age estimation were designed and developed to achieve high efficiency calculation of facial age estimation. In the first classifier, we adopt KNN-distance approach to calculate minimum distance between test face image and all instances belong to the class that has the highest number of nearest samples. Additionally, in the second classifier a modified-KNN version was proposed and the classifier scoring results interpolated to calculate the exact age estimation. Furthermore, KNN-regression classifier as third classifier that used to combine the classification and regression approaches to improve the accuracy of the age estimation system. Moreover, we compared age estimation errors under two situations: case 1, age estimation is performed without discrimination between males and females (gender unknown); and case 2, age estimation is performed for males and females separately (gender known). Results of experiments conducted on well know benchmark FG-NET Database show the effectiveness of the proposed approach.

A two-stage algorithm for extraction of ground and top of canopy in photon-counting profiling-LiDAR data in preparation for the ICESat-2 mission

Estimating forest canopy heights over large geographical regions is important for quantifying global carbon stocks, which in turn are needed for the study of climate change. Laser altimetry (LiDAR) is increasingly popular as a sensing technique for vegetation height and structure. NASAs ICESat-2, which is scheduled for launch in July 2017, will provide a platform for collecting LiDAR-based canopy height measurements on a global scale over a two-year period. However, unlike the first ICESat mission, which utilized a waveform LiDAR, ICESat-2 will be equipped with a photon-counting LiDAR system. This new sensor is capable of providing increased accuracy in high-slope terrain, but it is also expected to produce large amounts of noise, depending on sun angle and atmospheric conditions. This paper describes recent improvements to a novel two-stage algorithm that estimates ground and canopy surfaces from photon-counting LiDAR, in preparation for the ICESat-2 mission.

Content based image retrieval of diabetic macular edema images

Colour fundus images play an important role in diagnosing and screening diabetic macular edema (DME). In rural areas, content-based image retrieval (CBIR) might compensate the lack of expert ophthalmologists. In this work, we present a fully automated CBIR system that retrieves fundus images according to their content (quantity and location) of exudates. First, the macula is divided into three concentric regions whose texture discontinuities are used to represent lesion content of the retina. The image-to-image distance measure gives higher weights to lesions closer to the fovea to reflect the severity of the DME. Retrieval results of the system show precision of 79.2%.

Three-dimensional angiography using mobile C-arm with IMU sensor attached: Initial study

Three-dimensional (3D) computed tomography (CT) imaging is becoming an essential demand in several clinical procedures. Mobile C-arm is a useful imaging tool for image-guided interventional radiology. C-arm systems are provided with X-ray image intensifier (XRII) or flat-panel detectors. Essentially, C-arm CT systems requires scanners with flat-panel detectors for its ability to provide homogenous image quality and improve the resolution of low-contrast subjects compared to those equipped with XRII. However, C-arm systems with XRIIs are widely used in several interventional procedures. Such systems can provide a high quality two-dimensional (2D) fluoroscopic images that facilitates minimal invasive surgery. However, it is unable to provide depth information for 3D imaging due to several factors. First, the gantry of XRII-based C-arms is usually operated manually, where the rotation angle is determined using printed angle scale attached to the scanner gantry. Second, the gantry orbital rotation is normally limited to angular range less than theoretically required for exact 3D reconstruction. Third, considering the offset-scan geometry, which is common configuration in mobile C-arm with XRII, the number of rays passing through field-of-view (FOV) is limited. In this paper, we develop a 3D angiographic imaging system using commercial C-arm system equipped with XRII. First, an in-house made gantry rotation unit is developed to control the scanner orbital rotation. Second, the gantry rotation is traced using inertial measurement unit (IMU) sensor attached to the scanner gantry. Geometry information obtained from IMU sensor are used to define the gantry position in the 3D space and synchronized with detector measurements. The SCAN algorithm is used for the 3D reconstruction and achieved results are of high quality.

Content-based image retrieval strategies for medical image libraries

The objective of Content-based image retrieval (CBIR) in medical field is to permit radiologists to retrieve images of similar features that lead to similar diagnosis as the input image. This is different from other fields where the objective is to find the nearest images from the same category or match a part of an image. Therefore, such techniques cannot be directly applied in the medical field. In this study, a modified wavelet-based matching technique is introduced that is more robust to motion, noise, and brightness changes within the image. Also, we propose a description-based technique in which a semantic net is built in which each node represents a specific region and its spatial relation to other regions in the image. This semantic net can be considered as a hierarchical relationship tree with links between the nodes in the same level to describe the geographic relations between these nodes. Nodes contain region-specific features such as the moments of region boundaries in addition to local textural features. In the matching phase, the semantic net is built for the input image then used in the matching process. The matching process starts from the highest level in the hierarchical relationship tree for fast convergence.

Hybrid search based association rule mining

Association rule mining algorithms provide different search strategies to discover the frequent itemsets; however the problem of large search space is still hard to handle. The present paper suggests a novel hybrid search technique based on reducing the search space, I/O, and CPU times. This can improve the performance up to several orders of magnitude compared to APRIORI algorithm.

Integration of content-based image retrieval system with PACS

Content-based image retrieval (CBIR) provides a flexible means of searching a digital image library based on the description of the desired image. In this paper, we integrate CBIR, RIS, and HIS in PACS to allow retrieving images of similar features. Once the system finds the related images, the embedded CBIR retrieves the radiological reports and medical records of the output images, which can be used to increase diagnostic accuracy. The CBIR system is implemented on a separate server based on multi-resolution image matching. To reduce the retrieval loading on the server and network shanks, a procedure to use copies of images that are temporarily located in some workstations in the PACS is applied. These copies are stored on a temporary database space created on the different workstations. A new image retrieval management server contains image IDs in the database and the IP addresses of the workstations containing temporary image copies. Data on the management server are continuously updated with each addition or retrieval operation. When a display workstation needs a specific image, it sends a request with the required image ID to the management server, which in turn replies with the IP of the workstation containing the inquired image ID.

Automatically Human Age Estimation Approach via Two-Dimensional Facial Image Analysis

In this paper, the human age automatically estimated via two-dimensional facial image analysis. The exact age estimation is often treated as a classification problem while it can be formulated as a regression problem. In our research, a classification and regression models are proposed. The two proposed models are evaluated using the same database images and the same features. Due to a big difference between the number of samples in each class or age group, the two proposed models used the complete and missing data in different experiments. Moreover, we compared age estimation errors when (1) Age estimation is performed without discrimination between males and females (gender unknown); (2) Age estimation is performed in males and females separately (gender known). Conclusions and results of these proposed models are shown by extensive experiments on the public available FG-NET database.

Development of interactive 3D imaging system for hepatic angiography

Egypt is witnessing over the next decade many challenges in the field of healthcare, especially with regard to the spread of hepatitis and coincides with the spread of liver hepatocellular carcinoma. Because most cases of liver cancer in Egypt are detected in very late stages, the use of surgical resection, liver transplantation and percutaneous ablative therapies constitutes unsuitable therapeutic options either due to high recurrence rate or unfeasibility. Therapy sessions can be made through the introduction of chemotherapy using a catheter directly into the hepatic artery supplying the tumor guided by angiography imaging system. This method of treatment known to prevent the patient from different problems associated with surgical treatment, but it is still needs to be further improved to maximize the benefits and minimize the risks. Hepatic angiography is an x-ray study of the blood vessels that supply the liver. The procedure uses a catheter that is placed into a blood vessel through a small incision. The catheter is guided using the x-ray images obtained through the interventional session. During angiography, hepatic arterial supply is usually displayed in one, two or three projections. Mental 3D interpretation of the anatomy is not an easy task. Reaching the target supply artery by the catheter tip is mandatory to obtain satisfactory tumor response and reduce complications and recurrence. This work aims to develop an interactive 3D imaging system of hepatic angiography. The developed system uses a set of 2D images measured over few view angles to reconstruct a full 3D volume of the hepatic arteries. The problem can be thought as a combination of three main approaches. (1) Image reconstruction of 3D artery volume from few number of projections (each is presented as 2D image), (2) automatic detection of the catheter roadmap to the labeled artery which feed the tumor, and (3) interactive system to control and display images using simple gestures of the physician.