OBJECTIVES: Clinical discharge summaries provide valuable information about patients' clinical history, which is helpful for the realization of intelligent healthcare applications. The documents tend to take the form of separate segments based on temporal or topical information. If a patient's clinical history can be seen as a consecutive sequence of clinical events, then each temporal segment can be seen as a snapshot, providing a certain clinical context at a specific moment. This study aimed to demonstrate a temporal segmentation method of Korean clinical narratives for identifying textual snapshots of patient history as a proof-of-a-concept. METHODS: Our method uses pattern-based segmentation to approximate human recognition of the temporal or topical shifts in clinical documents. We utilized rheumatic patients' discharge summaries and transformed them into sequences of constituent chunks. We built 97 single pattern functions to denote whether a certain chunk has attributes that indicate that it can be a segment boundary. We manually defined the relationships between the pattern functions to resolve multiple pattern matchings and to make a final decision. RESULTS: The algorithm segmented 30 discharge summaries and processed 1,849 decision points. Three human judges were asked whether they agreed with the algorithm's prediction, and the agreement percentage on the judges' majority opinion was 89.61%. CONCLUSIONS: Although this method is based on manually constructed rules, our findings demonstrate that the proposed algorithm can achieve fairly good segmentation results, and it may be the basis for methodological improvement in the future.
Delivery of Health Care ; Electronic Health Records ; Humans ; Methods ; Natural Language Processing ; Pattern Recognition, Automated ; Rheumatic Diseases

Breast cancer is the second leading cancer for Korean women and its incidence rate has been increasing annually. If early diagnosis were implemented with epidemiologic data, the women could easily assess breast cancer risk using internet. National Cancer Institute in the United States has released a Web-based Breast Cancer Risk Assessment Tool based on Gail model. However, it is inapplicable directly to Korean women since breast cancer risk is dependent on race. Also, it shows low accuracy (58%-59%). In this study, breast cancer discrimination models for Korean women are developed using only epidemiological case-control data (n = 4,574). The models are configured by different classification techniques: support vector machine, artificial neural network, and Bayesian network. A 1,000-time repeated random sub-sampling validation is performed for diverse parameter conditions, respectively. The performance is evaluated and compared as an area under the receiver operating characteristic curve (AUC). According to age group and classification techniques, AUC, accuracy, sensitivity, specificity, and calculation time of all models were calculated and compared. Although the support vector machine took the longest calculation time, the highest classification performance has been achieved in the case of women older than 50 yr (AUC = 64%). The proposed model is dependent on demographic characteristics, reproductive factors, and lifestyle habits without using any clinical or genetic test. It is expected that the model could be implemented as a web-based discrimination tool for breast cancer. This tool can encourage potential breast cancer prone women to go the hospital for diagnostic tests.
Adult ; Aged ; Aged, 80 and over ; Breast Neoplasms/*diagnosis/*epidemiology ; Diagnosis, Computer-Assisted/*methods ; Early Detection of Cancer/*methods ; Female ; Humans ; *Machine Learning ; Middle Aged ; Pattern Recognition, Automated/methods ; Prevalence ; Reproducibility of Results ; Republic of Korea/epidemiology ; Risk Assessment/methods ; Risk Factors ; Sensitivity and Specificity ; Women's Health/*statistics & numerical data

The metabolomic analysis of three Cimicifuga species was performed using H-NMR spectroscopy and pattern recognition (PR) techniques. A broad range of metabolites could be detected by 'H-NMR spectroscopy without any chromatographic separation. The analysis using principal component analysis (PCA) and discriminant partial least square (DPLS) of the 1H-NMR spectrum showed a clear discrimination between C. foetida and the other two species. The major metabolites responsible for the discrimination were triterpenoid saponins and saccharides. These results indicated that the combination of 1H-NMR and PR provides a useful tool for chemotaxonomic analysis and authentification of Cimicifuga species, and could used for the quality control of plant materials.
Cimicifuga ; classification ; Discriminant Analysis ; Drugs, Chinese Herbal ; classification ; standards ; Magnetic Resonance Spectroscopy ; methods ; Metabolomics ; methods ; Pattern Recognition, Automated ; Principal Component Analysis ; Protons ; Saponins ; isolation & purification ; Triterpenes ; isolation & purification

To improve accuracy of estimation in planted safflower acreage,we selected agricultural area in Yumin County, Xinjiang as the study area. There safflower was concentrated planted. Supervised classification based on Principal Component Analysis (PCA) and texture feature were used to obtain the safflower acreage from image captured by ZY-3. The classification result was compared with only spectral feature and spectral feature with texture feature. The research result shows that this method can effectively solve the problem of low accuracy and fracture classification result in single data source classification. The overall accuracy is 87.519 1%, which increases by 7.117 2% compared with single data source classification. Therefore, the classification method based on PCA and texture features can be adapted to RS image classification and estimate the acreage of safflower. This study provides a feasible solution for estimation of planted safflower acreage by image captured by ZY-3 satellite.
Algorithms ; Carthamus tinctorius ; chemistry ; growth & development ; Image Processing, Computer-Assisted ; Pattern Recognition, Automated ; Principal Component Analysis ; methods ; Remote Sensing Technology ; methods

Facial paralysis is a frequently-occurring disease, which causes the loss of the voluntary muscles on one side of the face due to the damages the facial nerve and results in an inability to close the eye and leads to dropping of the angle of the mouth. There have been few objective methods to quantitatively diagnose it and assess this disease for clinically treating the patients so far. The skin temperature distribution of a healthy human body exhibits a contralateral symmetry. Facial paralysis usually causes an alteration of the temperature distribution of body with the disease. This paper presents the use of the histogram distance of bilateral local binary pattern (LBP) in the facial infrared thermography to measure the asymmetry degree of facial temperature distribution for objective assessing the severity of facial paralysis. Using this new method, we performed a controlled trial to assess the facial nerve function of the healthy subjects and the patients with Bell's palsy respectively. The results showed that the mean sensitivity and specificity of this method are 0.86 and 0.89 respectively. The correlation coefficient between the asymmetry degree of facial temperature distribution and the severity of facial paralysis is an average of 0.657. Therefore, the histogram distance of local binary pattern in the facial infrared thermography is an efficient clinical indicator with respect to the diagnosis and assessment of facial paralysis.
Facial Paralysis ; diagnosis ; physiopathology ; Humans ; Infrared Rays ; Pattern Recognition, Automated ; methods ; Skin Temperature ; Thermography ; instrumentation

Discriminative common vector (DCV) is an effective method that was proposed for the small sample size problems of face recognition. There is the same problem in brain-computer interface (BCI). Using directly the linear discriminative analysis (LDA) could result in errors because of the singularity of the within-class matrix of data. In our studies, we used the DCV method from the common vector theory in the within-class scatter matrix of data of all classes, and then applied eigenvalue decomposition to the common vectors to obtain the final projected vectors. Then we used kernel discriminative common vector (KDCV) with different kernel. Three data sets that include BCI Competition I data set, Competition II data set IV, and a data set collected by ourselves were used in the experiments. The experiment results of 93%, 77% and 97% showed that this feature extraction method could be used well in the classification of imagine data in BCI.
Algorithms ; Artificial Intelligence ; Brain-Computer Interfaces ; Discriminant Analysis ; Electroencephalography ; Face ; anatomy & histology ; Humans ; Pattern Recognition, Automated ; methods ; Principal Component Analysis ; Sample Size ; Signal Processing, Computer-Assisted ; User-Computer Interface

Medical image fusion and recognition has a wide range of applications, such as focal location, cancer staging and treatment effect assessment. Multi-model medical image fusion and recognition are analyzed and summarized in this paper. Firstly, the question of multi-model medical image fusion and recognition is discussed, and its advantage and key steps are discussed. Secondly, three fusion strategies are reviewed from the point of algorithm, and four fusion recognition structures are discussed. Thirdly, difficulties, challenges and possible future research direction are discussed.
Algorithms ; Artificial Intelligence ; Diagnostic Imaging ; methods ; Image Enhancement ; methods ; Image Interpretation, Computer-Assisted ; methods ; Image Processing, Computer-Assisted ; methods ; Magnetic Resonance Imaging ; methods ; Pattern Recognition, Automated ; methods ; Positron-Emission Tomography ; methods ; Tomography, X-Ray Computed ; methods

Due to the difficulties of the copying, vein identification has developed rapidly in recent years. The light source selection directly affects the image quality. This paper acquired by experiment the reflectivities of vein and non-vein irradiation with different wavelengths of near infra-red. Comparing the strength of reflectivities of various wave lengths, we found that there were the strongest contrasts between vein and non-vein in the 810 nm, and 810 nm near infra-red was suitable to a vein imaging light source. Finally, clear hand vein images were obtained with the selected light source.
Algorithms ; Biometry ; methods ; Hand ; anatomy & histology ; blood supply ; Humans ; Image Enhancement ; Image Processing, Computer-Assisted ; methods ; Infrared Rays ; Pattern Recognition, Automated ; Tomography, Optical ; methods ; Veins ; anatomy & histology

A fast and accurate intelligent identification is the developing trend of electrocardiogram (ECG) research. However, there are few methods by which satisfactory results could be obtained both in speed and in accuracy. A fast identification method of QRS wave was proposed based on moving window operation in this study. An 80 ms wide moving window was employed, in which simple difference and product operations were carried out with simple and less computation, and a very good inhibition of P and T waves and other noises was realized. Then the method was investigated with data from the MIT-BIH arrhythmia database with absence of digital filtering de-noising, and the identification accuracy of QRS complex reached 99.6%. The results showed that a rapid and accurate identification of QRS complex could be realized, which would meet the requirements for studying the real-time ECG monitoring equipment.
Algorithms ; Electrocardiography ; methods ; Humans ; Pattern Recognition, Automated ; Signal Processing, Computer-Assisted

To solve the ineffective problem of leukocytes classification based on multi-feature fusion in a single color space, we proposed an automatic leukocyte pattern recognition by means of feature fusion with color histogram and texture granular in multi-color space. The interactive performance of three color spaces (RGB, HSV and Lab), two features (color histogram and texture granular) and four similarity measured distance metrics (normalized intersection, Euclidean distance, chi2-metric distance and Mahalanobis distance) were discussed. The optimized classification modes of high precision, extensive universality and low cost to different leukocyte types were obtained respectively, and then the recognition system of tree-integration of the optimized modes was established. The experimental results proved that the performance of the fusion classification was improved by 12.3% at least.
Algorithms ; Clinical Laboratory Techniques ; Color ; Humans ; Image Enhancement ; methods ; Image Interpretation, Computer-Assisted ; methods ; Leukocyte Count ; methods ; Leukocytes ; classification ; cytology ; Pattern Recognition, Automated ; methods