模式识别在有机化合物碳—13核磁共振波谱中的应用

本文介绍了以碳—13 NMR 谱为基础,运用模式识别方法对于取代苯类有机化合物的分类情况。数据源为 CIAC-碳-13数据库。特征选择为简单的机率比率法。模式识别方法为Fisher 意义下的判别函数、KNN 及非线性映射。所得结果比较满意。

复杂背景下运动目标检测、分割与跟踪

本文研究的内容主要是序列图像中复杂背景下运动目标检测、分割与跟踪。分别提出了曲线演化策略及在运动目标检测与分割中的应用和基于特征点跟踪的运动目标检测与分割方法。关于目标跟踪，本文应用基于模型和基于特征的两种策略，提出了稳定、有效的四种跟踪方法。基于模型的算法包括基于粒子滤波和Hausdorff距离的运动目标跟踪及基于变形模板的目标跟踪。基于特征的方法分别是基于模糊测度的目标匹配与改进不变矩的目标匹配算法。 研究了曲线演化策略及在运动目标检测与分割中的应用，提出了基于最小半径准则和队列结构的聚类算法，消除孤立区域。提出基于试探策略并结合Fisher准则自适应地确定阈值的曲线演化算法分割运动目标。该算法具有良好的实时性和分割完整准确的优点。 在基于特征点跟踪的运动目标检测分割中，通过分析角点信息(灰度、梯度大小与方向)及其邻域信息，构造方向子邻域以提高匹配精度。提出了采用聚类方法消除运动目标上特征点对集合，同时也从剩余的特征点对集合中选取最优子集估计模型参数，检测运动目标。该算法可达到实时应用目的。 目标跟踪问题可看作状态空间中贝叶斯推理问题。在粒子滤波框架内，分别构造了跟踪过程的状态转移模型和测量模型，并提出了一种基于集合的模糊测度，实现了跟踪窗口随目标自适应变化、模板更新策略。对于刚体与非刚体的孤立目标，算法能够实现对目标的稳定跟踪。 在复杂场景下变形目标跟踪中，通过定义一种点到匹配点及其邻域的广义距离测度，增强了匹配的容错性。算法采用边缘相似点的距离均值和方差作为匹配的相似测度和置信准则。通过提出在8邻域内基于马尔可夫模型的启发式规则修正模板策略，阻止了模板窗口向背景的漂移。 在基于模糊测度的目标匹配中，提出了循环移位方法构造匹配特征向量，根据特征分布的模糊隶属度确定相似测度。该算法性能优于基于灰度相关匹配方法。 在基于不变矩的图像匹配中，引入误差带策略，计算图像边缘不变矩，取前N个匹配结果作为侯选目标，并根据契比雪夫不等式确定置信度。最后计算侯选目标的直方图的统计特性目标定位。该算法在性能上明显优于应用不变矩特征直接匹配方法和灰度相关的匹配方法。

基于模糊集理论的二维线性鉴别分析新方法

二维线性鉴别分析（2DLDA）是一种直接基于矩阵的特征提取方法，跳过传统的基于Fisher鉴别准则的线性鉴别分析方法中必须先将二维矩阵转化成一维矢量的过程，有效地提高了特征提取速度且避免了小样本问题，其识别率优于传统的Fisherface方法。结合模糊集理论，提出了一种新的2DLDA算法——模糊2DLDA（FIDLDA）算法。首先采用FKNN算法得到相应的样本分布信息，并按其对最后得到的特征向量所作的贡献融入到特征抽取过程中，得到有效的样本特征向量集。实验表明，F2DLDA算法的性能优于传统的2DLDA算法和Fisherface方法。

广义统计不相关最优鉴别矢量集的一个理论结果

本文对统计不相关最优鉴别矢量集的理论问题进行研究 ,提出了广义统计不相关最优鉴别准则 ,并给出了广义统计不相关最佳鉴别矢量集的一个理论结果 ,研究表明 ,广义统计不相关最佳鉴别矢量集的计算公式与基于Fisher最优鉴别准则的统计不相关最佳鉴别矢量集的计算公式完全一样 ,但是以前这一点没有被认识到 .本文的研究丰富了统计不相关最优鉴别分析的特征抽取理论 .

一种新的图像特征抽取方法研究

对最佳鉴别矢量的求解方法进行了研究，根据矩阵的分块理论和优化理论，在一定的条件下，从理论上得到类间散布矩阵和总体散布矩阵的一种简洁表示方法，提出了求解最佳鉴别矢量的一种新算法，该算法的优点是计算量明显减少。ORL人脸数据库的数值实验，验证了上述论断的正确性。实验结果表明，虽然识别率与分块维数之间存在非线性关系，但可以通过选择适当的分块维数来获得较高的识别率。类间散布矩阵和总体散布矩阵的一种简洁表示方法适合于一切使用Fisher鉴别准则的模式识别问题。

A Unified Statistical and Information Theoretic Framework for Multi-modal Image Registration

We formulate and interpret several multi-modal registration methods in the context of a unified statistical and information theoretic framework. A unified interpretation clarifies the implicit assumptions of each method yielding a better understanding of their relative strengths and weaknesses. Additionally, we discuss a generative statistical model from which we derive a novel analysis tool, the "auto-information function", as a means of assessing and exploiting the common spatial dependencies inherent in multi-modal imagery. We analytically derive useful properties of the "auto-information" as well as verify them empirically on multi-modal imagery. Among the useful aspects of the "auto-information function" is that it can be computed from imaging modalities independently and it allows one to decompose the search space of registration problems.

An artificial immune system for continuous analysis of time-varying data

M. Neal, An Artificial Immune System for Continuous Analysis of Time-Varying Data, in Proceedings of the 1st International Conference on Artificial Immune Systems (ICARIS), 2002, eds J Timmis and P J Bentley, volume 1, pages 76-85,

A resource limited artificial immune system for data analysis.

Timmis J and Neal M J. A resource limited artificial immune system for data analysis. In Proceedings of ES2000 - Research and Development of Intelligent Systems, pages 19-32, Cambrige, U.K., 2000. Springer.

Investigating the evolution and stability of a resource limited artificial immune system.

Timmis J and Neal M J. Investigating the evolution and stability of a resource limited artificial immune system. In Proceedings of GECCO - special workshop on artificial immune systems, pages 40-41. AAAI press, 2000.

Human papillomavirus detection in women with and without human immunodeficiency virus infection in Colombia

Background: HIV infection leads to a decreasing immune response, thereby facilitating the appearance of other infections, one of the most important ones being HPV. However, studies are needed for determining associations between immunodeficiency caused by HIV and/or the presence of HPV during the course of cervical lesions and their degree of malignancy. This study describes the cytological findings revealed by the Papanicolaou test, laboratory characteristics and HPV molecular profile in women with and without HIV infection. Methods: A total of 216 HIV-positive and 1,159 HIV-negative women were invited to participate in the study; PCR was used for the molecular detection of HPV in cervical samples. Statistical analysis (such as percentages, Chi-square test and Fisher's exact test when applicable) determined human papillomavirus (HPV) infection frequency (single and multiple) and the distribution of six types of high-risk-HPV in women with and without HIV infection. Likewise, a logistic regression model was run to evaluate the relationship between HIV-HPV infection and different risk factors. Results: An association was found between the frequency of HPV infection and infection involving 2 or more HPV types (also known as multiple HPV infection) in HIV-positive women (69.0% and 54.2%, respectively); such frequency was greater than that found in HIV-negative women (44.3% and 22.7%, respectively). Statistically significant differences were observed between both groups (p = 0.001) regarding HPV presence (both in infection and multiple HPV infection). HPV-16 was the most prevalent type in the population being studied (p = 0.001); other viral types had variable distribution in both groups (HIV-positive and HIV-negative). HPV detection was associated with <500 cell/mm(3) CD4-count (p = 0.004) and higher HIV-viral-load (p = 0.001). HPV-DNA detection, <200 cell/mm(3) CD4-count (p = 0.001), and higher HIV-viral-load (p = 0.001) were associated with abnormal cytological findings. Conclusions: The HIV-1 positive population in this study had high multiple HPV infection prevalence. The results for this population group also suggested a greater association between HPV-DNA presence and cytological findings. HPV detection, together with low CD4 count, could represent useful tools for identifying HIV-positive women at risk of developing cervical lesions.

A saúde oral em dentição infantil

Projeto de Pós-Graduação/Dissertação apresentado à Universidade Fernando Pessoa como parte dos requisitos para obtenção do grau de Mestre em Medicina Dentária

Prediction in dynamic models with time-dependent conditional variances

This paper considers forecasting the conditional mean and variance from a single-equation dynamic model with autocorrelated disturbances following an ARMA process, and innovations with time-dependent conditional heteroskedasticity as represented by a linear GARCH process. Expressions for the minimum MSE predictor and the conditional MSE are presented. We also derive the formula for all the theoretical moments of the prediction error distribution from a general dynamic model with GARCH(1, 1) innovations. These results are then used in the construction of ex ante prediction confidence intervals by means of the Cornish-Fisher asymptotic expansion. An empirical example relating to the uncertainty of the expected depreciation of foreign exchange rates illustrates the usefulness of the results. © 1992.

Ligament-derived matrix stimulates a ligamentous phenotype in human adipose-derived stem cells.

Human adipose stem cells (hASCs) can differentiate into a variety of phenotypes. Native extracellular matrix (e.g., demineralized bone matrix or small intestinal submucosa) can influence the growth and differentiation of stem cells. The hypothesis of this study was that a novel ligament-derived matrix (LDM) would enhance expression of a ligamentous phenotype in hASCs compared to collagen gel alone. LDM prepared using phosphate-buffered saline or 0.1% peracetic acid was mixed with collagen gel (COL) and was evaluated for its ability to induce proliferation, differentiation, and extracellular matrix synthesis in hASCs over 28 days in culture at different seeding densities (0, 0.25 x 10(6), 1 x 10(6), or 2 x 10(6) hASC/mL). Biochemical and gene expression data were analyzed using analysis of variance. Fisher's least significant difference test was used to determine differences between treatments following analysis of variance. hASCs in either LDM or COL demonstrated changes in gene expression consistent with ligament development. hASCs cultured with LDM demonstrated more dsDNA content, sulfated-glycosaminoglycan accumulation, and type I and III collagen synthesis, and released more sulfated-glycosaminoglycan and collagen into the medium compared to hASCs in COL (p

Unrecognized pretransplant and donor‐derived cryptococcal disease in organ transplant recipients.

BACKGROUND: Cryptococcosis occurring ≤30 days after transplantation is an unusual event, and its characteristics are not known. METHODS: Patients included 175 solid-organ transplant (SOT) recipients with cryptococcosis in a multicenter cohort. Very early-onset and late-onset cryptococcosis were defined as disease occurring ≤30 days or >30 days after transplantation, respectively. RESULTS: Very early-onset disease developed in 9 (5%) of the 175 patients at a mean of 5.7 days after transplantation. Overall, 55.6% (5 of 9) of the patients with very early-onset disease versus 25.9% (43 of 166) of the patients with late-onset disease were liver transplant recipients (P = .05). Very early cases were more likely to present with disease at unusual locations, including transplanted allograft and surgical fossa/site infections (55.6% vs 7.2%; P < .001). Two very early cases with onset on day 1 after transplantation (in a liver transplant recipient with Cryptococcus isolated from the lung and a heart transplant recipient with fungemia) likely were the result of undetected pretransplant disease. An additional 5 cases involving the allograft or surgical sites were likely the result of donor‐acquired infection. CONCLUSIONS: A subset of SOT recipients with cryptococcosis present very early after transplantation with disease that appears to occur preferentially in liver transplant recipients and involves unusual sites, such as the transplanted organ or the surgical site. These patients may have unrecognized pretransplant or donor-derived cryptococcosis.

Variation in use of surveillance colonoscopy among colorectal cancer survivors in the United States.

BACKGROUND: Clinical practice guidelines recommend colonoscopies at regular intervals for colorectal cancer (CRC) survivors. Using data from a large, multi-regional, population-based cohort, we describe the rate of surveillance colonoscopy and its association with geographic, sociodemographic, clinical, and health services characteristics. METHODS: We studied CRC survivors enrolled in the Cancer Care Outcomes Research and Surveillance (CanCORS) study. Eligible survivors were diagnosed between 2003 and 2005, had curative surgery for CRC, and were alive without recurrences 14 months after surgery with curative intent. Data came from patient interviews and medical record abstraction. We used a multivariate logit model to identify predictors of colonoscopy use. RESULTS: Despite guidelines recommending surveillance, only 49% of the 1423 eligible survivors received a colonoscopy within 14 months after surgery. We observed large regional differences (38% to 57%) across regions. Survivors who received screening colonoscopy were more likely to: have colon cancer than rectal cancer (OR = 1.41, 95% CI: 1.05-1.90); have visited a primary care physician (OR = 1.44, 95% CI: 1.14-1.82); and received adjuvant chemotherapy (OR = 1.75, 95% CI: 1.27-2.41). Compared to survivors with no comorbidities, survivors with moderate or severe comorbidities were less likely to receive surveillance colonoscopy (OR = 0.69, 95% CI: 0.49-0.98 and OR = 0.44, 95% CI: 0.29-0.66, respectively). CONCLUSIONS: Despite guidelines, more than half of CRC survivors did not receive surveillance colonoscopy within 14 months of surgery, with substantial variation by site of care. The association of primary care visits and adjuvant chemotherapy use suggests that access to care following surgery affects cancer surveillance.