Intelligent image analysis recognizes important orchid viral diseases.

Phalaenopsis orchids are one of the most important exporting commodities for Taiwan. Most orchids are planted and grown in greenhouses. Early detection of orchid diseases is crucially valuable to orchid farmers during orchid cultivation. At present, orchid viral diseases are generally identified with manual observation and the judgment of the grower's experience. The most commonly used assays for virus identification are nucleic acid amplification and serology. However, it is neither time nor cost efficient. Therefore, this study aimed to create a system for automatically identifying the common viral diseases in orchids using the orchid image. Our methods include the following steps: the image preprocessing by color space transformation and gamma correction, detection of leaves by a U-net model, removal of non-leaf fragment areas by connected component labeling, feature acquisition of leaf texture, and disease identification by the two-stage model with the integration of a random forest model and an inception network (deep learning) model. Thereby, the proposed system achieved the excellent accuracy of 0.9707 and 0.9180 for the image segmentation of orchid leaves and disease identification, respectively. Furthermore, this system outperformed the naked-eye identification for the easily misidentified categories [cymbidium mosaic virus (CymMV) and odontoglossum ringspot virus (ORSV)] with the accuracy of 0.842 using two-stage model and 0.667 by naked-eye identification. This system would benefit the orchid disease recognition for Phalaenopsis cultivation.

Construction of diagnosis system and gene regulatory networks based on microarray analysis.

A microarray analysis generally contains expression data of thousands of genes, but most of them are irrelevant to the disease of interest, making analyzing the genes concerning specific diseases complicated. Therefore, filtering out a few essential genes as well as their regulatory networks is critical, and a disease can be easily diagnosed just depending on the expression profiles of a few critical genes. In this study, a target gene screening (TGS) system, which is a microarray-based information system that integrates F-statistics, pattern recognition matching, a two-layer K-means classifier, a Parameter Detection Genetic Algorithm (PDGA), a genetic-based gene selector (GBG selector) and the association rule, was developed to screen out a small subset of genes that can discriminate malignant stages of cancers. During the first stage, F-statistic, pattern recognition matching, and a two-layer K-means classifier were applied in the system to filter out the 20 critical genes most relevant to ovarian cancer from 9600 genes, and the PDGA was used to decide the fittest values of the parameters for these critical genes. Among the 20 critical genes, 15 are associated with cancer progression. In the second stage, we further employed a GBG selector and the association rule to screen out seven target gene sets, each with only four to six genes, and each of which can precisely identify the malignancy stage of ovarian cancer based on their expression profiles. We further deduced the gene regulatory networks of the 20 critical genes by applying the Pearson correlation coefficient to evaluate the correlationship between the expression of each gene at the same stages and at different stages. Correlationships between gene pairs were calculated, and then, three regulatory networks were deduced. Their correlationships were further confirmed by the Ingenuity pathway analysis. The prognostic significances of the genes identified via regulatory networks were examined using online tools, and most represented biomarker candidates. In summary, our proposed system provides a new strategy to identify critical genes or biomarkers, as well as their regulatory networks, from microarray data.

Distance Metric Based Oversampling Method for Bioinformatics and Performance Evaluation.

An imbalanced classification means that a dataset has an unequal class distribution among its population. For any given dataset, regardless of any balancing issue, the predictions made by most classification methods are highly accurate for the majority class but significantly less accurate for the minority class. To overcome this problem, this study took several imbalanced datasets from the famed UCI datasets and designed and implemented an efficient algorithm which couples Top-N Reverse k-Nearest Neighbor (TRkNN) with the Synthetic Minority Oversampling TEchnique (SMOTE). The proposed algorithm was investigated by applying it to classification methods such as logistic regression (LR), C4.5, Support Vector Machine (SVM), and Back Propagation Neural Network (BPNN). This research also adopted different distance metrics to classify the same UCI datasets. The empirical results illustrate that the Euclidean and Manhattan distances are not only more accurate, but also show greater computational efficiency when compared to the Chebyshev and Cosine distances. Therefore, the proposed algorithm based on TRkNN and SMOTE can be widely used to handle imbalanced datasets. Our recommendations on choosing suitable distance metrics can also serve as a reference for future studies.

Blood Smear Image Based Malaria Parasite and Infected-Erythrocyte Detection and Segmentation.

In this study, an automatic malaria parasite detector is proposed to perceive the malaria-infected erythrocytes in a blood smear image and to separate parasites from the infected erythrocytes. The detector hence can verify whether a patient is infected with malaria. It could more objectively and efficiently help a doctor in diagnosing malaria. The experimental results show that the proposed method can provide impressive performance in segmenting the malaria-infected erythrocytes and the parasites from a blood smear image taken under a microscope. This paper also presents a weighted Sobel operation to compute the image gradient. The experimental results demonstrates that the weighted Sobel operation can provide more clear-cut and thinner object contours in object segmentation.

Factors associated with operative hysteroscopy outcome in patients with uterine adhesions or submucosal myomas.

OBJECTIVE: To determine the factors associated with hysteroscopic surgery long-term outcome in patients with intrauterine adhesions or submucosal myomas. METHODS: Factors thought to be associated with outcome were retrospectively evaluated from the records of 591 patients who were followed up for at least 5years after undergoing hysteroscopic adhesiolysis (n=203) or myomectomy (n=388). RESULTS: The major factors affecting outcome were degree of adhesion (OR, 1.91; P=0.03) in the former group and parity (OR, 0.55; P=0.005) and depth of intramural penetration of the myoma (OR, 30.74; P<0.001) in the latter. Severe intrauterine adhesion, low parity, and deep intramural penetration of submucosal myoma had an associated increase risk of poor outcome. The overall complication rate was 1.35% and, respectively, 12.8% and 9.3% of the patients who underwent hysteroscopic adhesiolysis or myomectomy needed a second intervention. CONCLUSION: Hysteroscopic surgery is a safe and effective procedure. Degree of adhesion or parity and depth of intramural penetration of myomas are the major factors affecting outcome in patients with these lesions.

A protozoan parasite extraction scheme for digital microscopic images.

Pathogenic protozoan parasites can cause human to get many diseases, such as, amoebiasis, typhoid fever and cholera, etc. Different protozoan parasites vary greatly in their structural and biochemical properties. Digital images are extensively applied to medical fields for doctors and pathologists to analyze pathological sections and further diagnose diseases. The aim of this paper is to develop protozoan parasite extraction techniques to segment protozoan parasites from microscopic images. The proposed scheme has precise segmentation ability even if the image is with poor quality or complex background. Experimental results show that the proposed scheme can gain 96.64% average correct rate, and about 0.04, 0.45 and 0.06 of the average error rates: misclassification error (ME), region non-uniformity (RN) and relative foreground area error (RFAE), respectively.