Parasitic worms are significant causes of human and livestock disease. The battle against infections caused by parasitic worms involves the exploration of numerous potential drug candidates. One approach in screening for new drug candidates is using natural product extracts on the nematode C. elegans as a model organism. A critical step in this process is the examination of microscopy images of C. elegans after exposure to natural product extracts. Automatic image classification accelerates the analysis process compared to purely visual identification by an expert. We report a new C. elegans image dataset including 12 717 microscopy images corresponding to natural product extracts, with about one-third of the images labeled by an expert and the remaining unlabeled. We make this dataset available to researchers for further development. We also propose a two-stage Semi-supervised Mix-up Barlow Twins Nematode Classifier (MBT-NC) to solve three image classification tasks involving nematode phenotypes after exposure to the studied natural extracts. MBT-NC combines self-supervised learning (SSL) for the feature representation stage (MBT) with a supervised classification stage (NC). In MBT, we utilize augmented and linearly interpolated samples for information maximization. Our method outperforms fully supervised and also other self-supervised methods on all three classification tasks: For binary, six-class, and 27-class classification, we outperform by 3.2%, 1.0%, and 2.2% respectively on test accuracy compared to the other methods. This is a new line of research in computer vision applications in healthcare.

In this paper we present a new method of uncovering patients with aortic valve diseases in large electronic health record systems through learning with multimodal data. The method automatically extracts clinically-relevant valvular disease features from five multimodal sources of information including structured diagnosis, echocardiogram reports, and echocardiogram imaging studies. It combines these partial evidence features in a random forests learning framework to predict patients likely to have the disease. Results of a retrospective clinical study from a 1000 patient dataset are presented that indicate that over 25 % new patients with moderate to severe aortic stenosis can be automatically discovered by our method that were previously missed from the records. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

This work reports the use of ultrasound radio frequency (RF) time series analysis as a method for ultrasound-based classification of malignant breast lesions. The RF time series method is versatile and requires only a few seconds of raw ultrasound data with no need for additional instrumentation. Using the RF time series features, and a machine learning framework, we have generated malignancy maps, from the estimated cancer likelihood, for decision support in biopsy recommendation. These maps depict the likelihood of malignancy for regions of size 1 mm(2) within the suspicious lesions. We report an area under receiver operating characteristics curve of 0.86 (95% confidence interval [CI]: 0.84%-0.90%) using support vector machines and 0.81 (95% CI: 0.78-0.85) using Random Forests classification algorithms, on 22 subjects with leave-one-subject-out cross-validation. Changing the classification method yielded consistent results which indicates the robustness of this tissue typing method. The findings of this report suggest that ultrasound RF time series, along with the developed machine learning framework, can help in differentiating malignant from benign breast lesions, subsequently reducing the number of unnecessary biopsies after mammography screening.

We aim to compute the movement of permanent stranded implant brachytherapy radioactive sources (seeds) in the prostate from the planned seed distribution to the intraoperative fluoroscopic distribution, and then to the postimplant computed tomography (CT) distribution. We present a novel approach to matching the seeds in these distributions to the plan by grouping the seeds into needle tracks. First, we identify the implantation axis using a sample consensus algorithm. Then, we use a network flow algorithm to group seeds into their needle tracks. Finally, we match the needles from the three stages using both their transverse plane location and the number of seeds per needle. We validated our approach on eight clinical prostate brachytherapy cases, having a total of 871 brachytherapy seeds distributed in 193 needles. For the intraoperative and postimplant data, 99.31% and 99.41% of the seeds were correctly assigned, respectively. For both the preplan to fluoroscopic and fluoroscopic to CT registrations, 100% of the needles were correctly matched. We show that there is an average intraoperative seed displacement of 4.94±2.42 mm and a further 2.97±1.81 mm of postimplant movement. This information reveals several directional trends and can be used for quality control, treatment planning, and intraoperative dosimetry that fuses ultrasound and fluoroscopy.

Prostate segmentation in B-mode images is a challenging task even when done manually by experts. In this paper we propose a 3D automatic prostate segmentation algorithm which makes use of information from both ultrasound B-mode and vibro-elastography data.We exploit the high contrast to noise ratio of vibro-elastography images of the prostate, in addition to the commonly used B-mode images, to implement a 2D Active Shape Model (ASM)-based segmentation algorithm on the midgland image. The prostate model is deformed by a combination of two measures: the gray level similarity and the continuity of the prostate edge in both image types. The automatically obtained mid-gland contour is then used to initialize a 3D segmentation algorithm which models the prostate as a tapered and warped ellipsoid. Vibro-elastography images are used in addition to ultrasound images to improve boundary detection.We report a Dice similarity coefficient of 0.87±0.07 and 0.87±0.08 comparing the 2D automatic contours with manual contours of two observers on 61 images. For 11 cases, a whole gland volume error of 10.2±2.2% and 13.5±4.1% and whole gland volume difference of -7.2±9.1% and -13.3±12.6% between 3D automatic and manual surfaces of two observers is obtained. This is the first validated work showing the fusion of B-mode and vibro-elastography data for automatic 3D segmentation of the prostate.

Many genetic disorders or possible abnormalities that may occur in the future generations can be predicted through analyzing the shape and morphological characteristics of the chromosomes. Karyotype (a systemized array of the human chromosomes obtained from a single cell either by drawing or by photography using a light microscope is often used for this purpose. To make a Karyotype it is necessary to identify each one of the 24 chromosomes (22 autosomal and a pair of sex chromosomes) from the microscopic images. The first step to automate this process is then to define the morphological and band pattern based features for each chromosome. An important class of morphological features includes those defined with respect to the location of the chromosome's centromere (part of the chromosome that divides it to the long and short arms). Therefore, localization of centromere is an initial step in designing an automatic karyotyping system. In this paper, an effective algorithm for chromosome image processing and automatic centromere locating is presented The procedure is based on the calculation and analyzing the vertical and horizontal projection vectors of the binary image of the chromosome. The binary image is obtained using the thresholding of the input image after histogram modification and analyzing. When applied to the real chromosome images supplied by the Cytogenetic Laboratory of the Cancer Institute of the Imam hospital in Tehran, an average accuracy of 96% for Centromere locating is achieved.