In dentistry, radiological examinations help specialists by showing structure of the tooth bones with the goal of screening embedded teeth, bone abnormalities, cysts, tumors, infections, fractures, problems in the temporomandibular regions, just to cite a few. Sometimes, relying solely in the specialist's opinion can bring differences in the diagnoses, which can ultimately hinder the treatment. Although tools for complete automatic diagnosis are no yet expected, image pattern recognition has evolved towards decision support, mainly starting with the detection of teeth and their components in X-ray images. Tooth detection has been object of research during at least the last two decades, mainly relying in threshold and region-based methods. Following a different direction, this paper proposes to explore a deep learning method for instance segmentation of the teeth. To the best of our knowledge, it is the first system that detects and segment each tooth in panoramic X-ray images. It is noteworthy that this image type is the most challenging one to isolate teeth, since it shows other parts of patient's body (e.g., chin, spine and jaws). We propose a segmentation system based on mask region-based convolutional neural network to accomplish an instance segmentation. Performance was thoroughly assessed from a 1500 challenging image data set, with high variation and containing 10 categories of different types of buccal image. By training the proposed system with only 193 images of mouth containing 32 teeth in average, using transfer learning strategies, we achieved 98% of accuracy, 88% of F1-score, 94% of precision, 84% of recall and 99% of specificity over 1224 unseen images, results very superior than other 10 unsupervised methods.

Deep Instance Segmentation of Teeth in Panoramic X-Ray Images

This review presents an in-depth study of the literature on segmentation methods applied in dental imaging. Several works on dental image segmentation were studied and categorized according to the type of method (region-based, threshold-based, cluster-based, boundary-based or watershed-based), type of X-ray images analyzed (intra-oral or extra-oral), and characteristics of the data set used to evaluate the methods in each state-of-the-art work. We found that the literature has primarily focused on threshold-based segmentation methods (54%). 80% of the reviewed articles have used intra-oral X-ray images in their experiments, demonstrating preference to perform segmentation on images of already isolated parts of the teeth, rather than using extra-oral X-rays, which also show tooth structure of the mouth and bones of the face. To fill a scientific gap in the field, a novel data set based on extra-oral X-ray images, presenting high variability and with a large number of images, is introduced here. A statistical comparison of the results of 10 pixel-wise image segmentation methods over our proposed data set comprised of 1500 images is also carried out, providing a comprehensive source of performance assessment. Discussion on limitations of the benchmarked methods, as well as future perspectives on exploiting learning-based segmentation methods to improve performance, is also addressed. Finally, we present a preliminary application of the MASK recurrent convolutional neural network to demonstrate the power of a deep learning method to segment images from our data set.

Automatic segmenting teeth in X-ray images: Trends, a novel data set, benchmarking and future perspectives

In this paper, we propose a classification approach for ambiguously shaped defects found on the surface of a type of display panel module that is widely used in the field of mobile displays. These types of surface defects are difficult to properly distinguish due to defect similarity and diversity. In such cases, defect types can only be determined using cumbersome human visual inspection. To solve the problem of ambiguous surface defect classification, we introduce a novel filtering method that effectively separates the foreground defective regions from the background, which has structured patterns, local illumination variation, and different light conditions for each of several cameras in an inspection system. Applying the proposed filter method to defect images, we select important features by adopting a wrapper-based feature selection method using a random forest as a learning algorithm. Successful classification results using the presented model are obtained using challenging real-world defect image data gathered from a smart phone display module inspection line in an industrial plant.

Ambiguous Surface Defect Image Classification of AMOLED Displays in Smartphones

Multimodal sentiments have become the challenge for the researchers and are equally sophisticated for an appliance to understand. One of the studies that support MS problems is a MSA, which is the training of emotions, attitude, and opinion from the audiovisual format. This survey article covers the comprehensive overview of the last update in this field. Many recently proposed algorithms and various MSA applications are presented briefly in this survey. The article is categorized according to their contributions in the various MSA techniques. The main purpose of this survey is to provide a full image of the MSA opportunities and difficulties and related field with brief details. The main contribution of this article includes the sophisticated categorizations of a large number of recent articles and the illustration of the recent trend of research in the MSA and its related areas.

Multimodal Sentiment Analysis: A Survey and Comparison

Despite the increasingly digital nature of society there are some areas of research that remain firmly rooted in the past; in this case the laboratory notebook, the last remaining paper component of an experiment Countless electronic laboratory notebooks (ELNs) have been created in an attempt to digitise record keeping processes in the lab, but none of them have become a ‘key player’ in the ELN market, due to the many adoption barriers that have been identified in previous research and further explored in the user studies presented here The main issues identified are the cost of the current available ELNs, their ease of use (or lack of it) and their accessibility issues across different devices and operating systems Evidence suggests that whilst scientists willingly make use of generic notebooking software, spreadsheets and other general office and scientific tools to aid their work, current ELNs are lacking in the required functionality to meet the needs of the researchers In this paper we present our extensive research and user study results to propose an ELN built upon a pre-existing cloud notebook platform that makes use of accessible popular scientific software and semantic web technologies to help overcome the identified barriers to adoption

/pdf/electronic-lab-notebooks-can-they-replace-paper-2cmpr3egf1.pdf

Electronic lab notebooks: can they replace paper?

Adaptive threshold works at pixel level and the result of the adaptive threshold is either background or foreground. The adaptive threshold produces superior result compared to global threshold, especially for the images that have uneven pixel intensity distribution. In the dental age assessment, X-ray image is used as an aid to estimate the age of the person. The existing process of assessment is done manually. However, this can be made automatically. The process of automated dental age assessment, require threshold segmentation to separate the background and the teeth area. In order to optimize the result of the adaptive threshold, it depends on the threshold value. In this paper, we present three methods (i.e. mean, median and OTSU) to estimate the range of the threshold value. The result of the study shown that the median threshold provides better results than the mean and OTSU thresholds. In terms of the region of the segmentation, median threshold value covers more teeth followed by mean threshold and OTSU threshold. The region of segmentation is important because one of the requirements in Demirjian method is to assess all the teeth types in quadrant 2 and quadrant 3.Based on the result of the experiment shown the region of median threshold able to segment most of the teeth area in quadrant 2 and quadrant 3.

Region of adaptive threshold segmentation between mean, median and otsu threshold for dental age assessment

The x-ray image is a grey scale image and the distribution of the intensity of the pixel is uneven. The x-ray image widely use in dental age assessment especially Demirjian method. The purpose of the dental age assessment is to estimate the age of unidentified bodies. The current process is done manually by the examiner. The process potentially converted to an automated system. The development an automated dental age assessment required segmentation process, which is dividing the image into multiple meaningful parts based on region and edge. The edge segmentation form a contour based on the links detected. The authors present two types of edge segmentation methods (i.e. Sobel and Canny). The objective of the study is to make a comparison between the two methods. Result showed Sobel method was able to segment all the teeth area and remove the noise on the x-ray image while Canny algorithm was not able to segment all the teeth area especially incisors. The region of segmentation is important because one of the requirements in Demirjian method is to assess all the teeth types in quadrant 2 and quadrant 3. Based on the result, the experiment showed the Sobel algorithm able to segment most of the teeth area in quadrant 2 and quadrant 3.

Sobel and Canny Edges Segmentations for the Dental Age Assessment

The scatter form of multimedia data such as text, image, audio, and video posted regularly in the social media may contain useful information for the organizations. But, this information should be derived with the use of some form of analysis known as Multimodal Sentiment Analysis (MSA). But, there is a lack of proper analytic tools for such analysis. This paper presents a thorough overview of more than fifty most recent MSA research articles to find the gaps in terms of tasks, approaches theories and applications used till date. There seems to be no single approach, theory, and tool which can support MSA. The study showed that each and every mode presents different difficulties which have not bee n fully solved yet, such as feature points of a face, voice clarity in audio, video summarization and so on, and are great research opportunities for the future researchers. Also, this research recommends a list of existing and upcoming difficulties and opportunities of MSA research.

An overview of Multimodal Sentiment Analysis research: Opportunities and Difficulties

Architecture design of a user-orientated electronic laboratory notebook: A case study within an atmospheric chemistry community

Qiraat mobile has been gaining a lot of attention from users who wish to recite Quran. In this paper, we will discuss the earlier versions of MyQiraat derived from two previous projects of Qiraat mobile design called MyQiraat version 1 and MyQiraat version 2. The aim of this research is to enable and promote the learning of Qiraat anytime and anywhere using mobile devices. The initial evaluation result shows supportive response from the users who wish to learn Qiraat via mobile devices provided the users' feedback is considered in the enhancement version.

Zulkifly Mohd Zaki

Papers

Region of adaptive threshold segmentation between mean, median and otsu threshold for dental age assessment

Sobel and Canny Edges Segmentations for the Dental Age Assessment

An overview of Multimodal Sentiment Analysis research: Opportunities and Difficulties

Architecture design of a user-orientated electronic laboratory notebook: A case study within an atmospheric chemistry community

MyQiraat: An interactive Qiraat mobile application