Search

Background: Autism spectrum disorder (ASD) currently affects nearly 1 in 160 children worldwide. In over two-thirds of evaluations, no validated diagnostics are used and gold standard diagnostic tools are used in less than 5% of evaluations. Currently, the diagnosis of ASD requires lengthy and expensive tests, in addition to clinical confirmation. Therefore, fast, cheap, portable, and easy-to-administer screening instruments for ASD are required. Several studies have shown that children with ASD have a lower preference for social scenes compared with children without ASD. Based on this, eye-tracking and measurement of gaze preference for social scenes has been used as a screening tool for ASD. Currently available eye-tracking software requires intensive calibration, training, or holding of the head to prevent interference with gaze recognition limiting its use in children with ASD., Methods: In this study, we designed a simple eye-tracking algorithm that does not require calibration or head holding, as a platform for future validation of a cost-effective ASD potential screening instrument. This system operates on a portable and inexpensive tablet to measure gaze preference of children for social compared to abstract scenes. A child watches a one-minute stimulus video composed of a social scene projected on the left side and an abstract scene projected on the right side of the tablet's screen. We designed five stimulus videos by changing the social/abstract scenes. Every child observed all the five videos in random order. We developed an eye-tracking algorithm that calculates the child's gaze preference for the social and abstract scenes, estimated as the percentage of the accumulated time that the child observes the left or right side of the screen, respectively. Twenty-three children without a prior history of ASD and 8 children with a clinical diagnosis of ASD were evaluated. The recorded video of the child´s eye movement was analyzed both manually by an observer and automatically by our algorithm., Results: This study demonstrates that the algorithm correctly differentiates visual preference for either the left or right side of the screen (social or abstract scenes), identifies distractions, and maintains high accuracy compared to the manual classification. The error of the algorithm was 1.52%, when compared to the gold standard of manual observation., Discussion: This tablet-based gaze preference/eye-tracking algorithm can estimate gaze preference in both children with ASD and without ASD to a high degree of accuracy, without the need for calibration, training, or restraint of the children. This system can be utilized in low-resource settings as a portable and cost-effective potential screening tool for ASD.

Parasitic infections are generally diagnosed by professionals trained to recognize the morphological characteristics of the eggs in microscopic images of fecal smears. However, this laboratory diagnosis requires medical specialists which are lacking in many of the areas where these infections are most prevalent. In response to this public health issue, we developed a software based on pattern recognition analysis from microscopi digital images of fecal smears, capable of automatically recognizing and diagnosing common human intestinal parasites. To this end, we selected 229, 124, 217, and 229 objects from microscopic images of fecal smears positive for Taenia sp., Trichuris trichiura, Diphyllobothrium latum, and Fasciola hepatica, respectively. Representative photographs were selected by a parasitologist. We then implemented our algorithm in the open source program SCILAB. The algorithm processes the image by first converting to gray-scale, then applies a fourteen step filtering process, and produces a skeletonized and tri-colored image. The features extracted fall into two general categories: geometric characteristics and brightness descriptions. Individual characteristics were quantified and evaluated with a logistic regression to model their ability to correctly identify each parasite separately. Subsequently, all algorithms were evaluated for false positive cross reactivity with the other parasites studied, excepting Taenia sp. which shares very few morphological characteristics with the others. The principal result showed that our algorithm reached sensitivities between 99.10%-100% and specificities between 98.13%- 98.38% to detect each parasite separately. We did not find any cross-positivity in the algorithms for the three parasites evaluated. In conclusion, the results demonstrated the capacity of our computer algorithm to automatically recognize and diagnose Taenia sp., Trichuris trichiura, Diphyllobothrium latum, and Fasciola hepatica with a high sensitivity and specificity.

Pneumonia is a disease which causes high mortality in children under five years old, particularly in developing countries. This paper proposes a novel application of ultrasound video analysis for the detection of pneumonia. This application is based on the processing of small video chunks, in which an image processing algorithm analyzes each frame to get some overall video statistics. Then, based on these quantities, the likeness of presence of pneumonia in the video is determined. The algorithm exploits different geometrical properties of typical anatomical and pathological features that commonly appear in lung sonography and which are already clinically typified in the literature. Our technique has been tested on different transverse thoracic scanning protocols and probe's maneuvers, thus, under a variety of clinical and usage protocols. Then, it can be targeted towards screening applications. We present encouraging results (AUC measure between 0.7851 and 0.9177) based on the analysis of 346 videos with an average duration of eight seconds. The analyzed videos were taken from children who were between three and five years old. Finally, our algorithm can be used directly as a classifier, but we detail how its performance may be enhanced if used as a first stage of a larger pipeline of other complementary pneumonia detection processes.

This work proposes an algorithm aimed at recognizing and accounting Koch bacilli in digital images of microbiological sputum samples stained with auramine, in order to determine the degree of concentration and the state of the disease (tuberculosis). The algorithm was developed with the main objective of maximizing the sensitivity and specificity of the analysis of microbiological samples (recognition and counting of bacilli) according to each preparation method (direct and diluted pellets) in order to reduce the subjectivity of the visual inspection applied by the specialist at the time of analyzing the samples. The proposed algorithm consists of a background removal, an image improvement stage based on consecutive morphological closing operations, a segmentation stage of objects of interest based on thresholdization and a classification stage based on SVM. Each algorithmic stage was developed taking into account the method of preparation of the sample to be processed, being this aspect the main contribution of the proposed work, since it was possible to achieve very satisfactory results in terms of specificity and sensitivity. In this context, sensitivity levels of 91.24% and 93.79% were obtained. Specificity levels of 90.33% and 94.85% were also achieved for direct and diluted pellet methods respectively. [ABSTRACT FROM AUTHOR]

This work proposes an algorithm oriented to the detection of tuberculosis bacilli in digital images of sputum samples, inked with the Ziehl Neelsen method and prepared with the direct, pellet and diluted pellet methods. The algorithm aims at automating the optical analysis of bacilli count and the calculation of the concentration level. Several algorithms have been proposed in the literature with the same objective, however, in no case is the performance in sensitivity and specificity evaluated for the 3 preparation methods. The proposed algorithm improves the contrast of the colors of interest, then thresholds the image and segments by labeling the objects of interest (bacilli). Each object then has its geometrical descriptors and photometric descriptors. With all this, a characteristic vector is formed, which are used in the training and classification process of an SVM. For the training 225 images obtained by the 3 preparation methods were used. The proposed algorithm reached, for the direct method, a sensitivity level of 93.67% and a specificity level of 89.23%. In the case of the Pellet method, a sensitivity of 92.13% and a specificity of 82.58% was obtained, while for diluted Pellet the sensitivity was 92.81% and the specificity 83.61%. [ABSTRACT FROM AUTHOR]

Pneumonia is one of the major causes of child mortality, yet with a timely diagnosis, it is usually curable with antibiotic therapy. In many developing regions, diagnosing pneumonia remains a challenge, due to shortages of medical resources. Lung ultrasound has proved to be a useful tool to detect lung consolidation as evidence of pneumonia. However, diagnosis of pneumonia by ultrasound has limitations: it is operator-dependent, and it needs to be carried out and interpreted by trained personnel. Pattern recognition and image analysis is a potential tool to enable automatic diagnosis of pneumonia consolidation without requiring an expert analyst. This paper presents a method for automatic classification of pneumonia using ultrasound imaging of the lungs and pattern recognition. The approach presented here is based on the analysis of brightness distribution patterns present in rectangular segments (here called “characteristic vectors“) from the ultrasound digital images. In a first step we identified and eliminated the skin and subcutaneous tissue (fat and muscle) in lung ultrasound frames, and the “characteristic vectors”were analyzed using standard neural networks using artificial intelligence methods. We analyzed 60 lung ultrasound frames corresponding to 21 children under age 5 years (15 children with confirmed pneumonia by clinical examination and X-rays, and 6 children with no pulmonary disease) from a hospital based population in Lima, Peru. Lung ultrasound images were obtained using an Ultrasonix ultrasound device. A total of 1450 positive (pneumonia) and 1605 negative (normal lung) vectors were analyzed with standard neural networks, and used to create an algorithm to differentiate lung infiltrates from healthy lung. A neural network was trained using the algorithm and it was able to correctly identify pneumonia infiltrates, with 90.9% sensitivity and 100% specificity. This approach may be used to develop operator-independent computer algorithms for pneumonia diagnosis using ultrasound in young children. [ABSTRACT FROM AUTHOR]