胰腺癌相关糖尿病的血清蛋白质组学分析

目的 检测胰腺癌相关糖尿病的血清蛋白标志物,并建立诊断模型.方法 应用表面增强激光解析电离飞行时间质谱( SELDI-TOF-MS)技术检测17例胰腺癌相关糖尿病与17例新发2型糖尿病、17例健康对照者血清的差异表达蛋白,用Biomarker Patterns Software 5.0软件建立胰腺癌相关糖尿病诊断模型并验证.结果 在胰腺癌相关糖尿病、新发2型糖尿病,健康者各10例的蛋白指纹图谱中筛选出12个差异表达蛋白峰,其中质荷比为6116、6695、8936 Da的蛋白峰被选为建立胰腺癌相关糖尿病诊断模型的蛋白峰.该诊断模型的诊断正确率为90％.盲法验证各组另7例样本,正确诊断胰腺癌相关糖尿病患者达100％,新发2型糖尿病患者为71％,健康人群为86％.经检索蛋白质数据库,与以上3种差异表达蛋白分子质量最为接近的蛋白分别为金属硫蛋白、胰腺干细胞增殖分化因子和成纤维细胞生长因子 1.结论 通过SELDI方法筛选出3种胰腺癌相关糖尿病的血清蛋白标志物,建立了可靠的胰腺癌相关糖尿病的诊断模型.     

SELDI技术筛选肺癌患者血清标志蛋白的临床价值

目的探讨表面增强激光解析电离飞行时间质谱（SELDI-TOF-MS）技术筛选肺癌患者血清标志蛋白的临床价值。方法用SELDI-TOF-MS技术、弱阳离子交换蛋白芯片,检测肺癌和肺良性病变患者的血清蛋白质质谱图;用Biomarker Pattern软件分析肺癌差异蛋白并初建其诊断模型,通过盲筛验证诊断模型。结果发现有统计学差异的蛋白峰20个,其中肺癌患者血清高表达蛋白质波峰14个,低表达蛋白质波峰6个;用质荷比2 090.77、2 503.31 Da的差异蛋白峰建立分类树模型,其诊断肺癌的灵敏度88%,特异度95%;盲筛验证灵敏度90%,特异度100%,粗符合率93.33%,Youden指数0.9。结论SELDI-TOF-MS技术筛选的肺癌血清差异性蛋白及分类树模型,诊断肺癌的灵敏度高、特异性好。     

2型糖尿病肾病尿液蛋白质谱及诊断决策树模型的建立

收集72例2型糖尿病肾病患者、33例2型糖尿病不伴肾病患者和29例正常人晨尿上清，采用铜离子螯合表面芯片（IMAC3-Cu^2＋），经表面加强激光解吸电离-飞行时间质谱测定得到蛋白质指纹图谱。建立了诊断决策树模型。糖尿病肾病组与非糖尿病肾病组相比较，17个蛋白峰强度相差2倍以上且有统计学意义（P〈0．01），其中4个蛋白建立了DN诊断模型，盲法验证该模型敏感性86．67％（26／30），特异性85．00％（17／20）。     

应用SELDI-TOF-MS技术建立肝癌筛选血清蛋白质指纹图谱模型

目的:建立肝癌筛选血清蛋白质指纹图谱模型．方法:用表面加强激光解析电离飞行时间质谱技术(SELDI-TOF-MS)及WCX2蛋白芯片获得新发肝癌、肝硬化患者和正常人血清的蛋白质指纹图谱,用计算机软件进行比较分析,建立肝癌的筛选模型．结果:肝癌患者与健康对照组血清蛋白质指纹图谱之间有5个标志蛋白(4477,8943,5181, 8617,13 761 Da)在肝癌患者血清中高表达,肝癌患者与肝硬化患者血清蛋白质指纹图谱之间2个标志蛋白(4477,13 761 Da)在肝癌患者血清中高表达,1个标志蛋白(4097 Da)在肝癌患者血清中低表达．SELDI-TOF-MS技术的特异性(60／60,100％);敏感度(18／20,90％)．分析系统筛选出4477,8943,13 761,4097 Da标志蛋白建立的肝癌诊断模型．结论:建立的血清蛋白质指纹图谱模型能够区分肝癌与非肝癌患者,SELDI-TOF-MS在肝癌的诊断及肿瘤特异性蛋白质生物标志分子的筛选等方面具有一定价值．     

血清肿瘤标志物DR-70^TM对原发性肝癌的诊断价值

目的评价血清肿瘤标志物DR-70^TM在原发性肝癌中的诊断价值，提高原发性肝癌（PHC）的检出率。方法应用美国AMDL公司的DR-70^TMELISA试剂盒，对20例正常人和60例AFP阳性原发性肝癌、43例AFP阴性原发性肝癌及30例肝硬化病人血清进行检测DR-70含量。结果原发性肝癌组血清DR-70^TM含量明显高于肝硬化组和正常对照组（P均〈0．001），AFP阳性原发性肝癌DR-70的阳性率为81．67％，AFP阴性原发性肝癌为93．02％，差异有显著性（P〈0．05）。DR-70^TM诊断肝癌敏感性为86．4％，特异性为94％；肝硬化组阳性率10％，特异性90％。肝硬化组与正常组比较差异无显著性（P〉0．05）；DR-70^TM与癌肿（直径）大小无明显关系（P〉0．05）。结论血清DR-70和AFP联合检测可大大提高原发性肝癌的阳性率，有利于原发性肝癌的早期诊断，特别是AFP阴性的肝癌的有价值的标志物。也有利于PHC与肝硬化（LC）的鉴别诊断。对非恶性肿瘤病人特异性较高，对人群筛选将有较高的参考价值。     

肝癌组织谷氨酰转移酶GGTmRNA-H亚型表达与肝癌早期诊断的研究

目的 研究肝癌组织中谷氨酰转移酶GGTmRNA—H亚型表达，并探讨其在肝癌早期诊断中的意义。方法 应用RT—PCR法，检测65例原发性肝癌病人癌组织及癌周组织、108例良性肝病肝组织、6例正常对照肝组织的GGTmRNA—H亚型，观察该基因的表达情况。结果 在65例原发性肝癌中，癌组织GGTmRNA—H亚型的阳性例数为61例（93．8％），肝组织血清AFP阳性40例（61，5％）（P〈0．05）；GGTmRNA—H亚型在20例小肝癌癌组织阳性表达例数为18例（阳性率为90．0％），与正常对照组比较，差异有统计学意义。急慢性肝炎组、肝硬化组与正常对照组比较，GGTmRNAH亚型阳性表达率虽明显升高，但其差异无统计学意义（P〉0．05）。结论 肝癌组织中（如TmRNA—H亚型与血AFP联合分析可弥补AFP检测的不足，对提高肝癌、癌前病变的早期诊断具有重要意义。GGTmRNA—H亚型基因检测的结果比AFP的检测更敏感、特异性更高，有望成为肝癌早期诊断的理想指标。     

原发性肝细胞癌组织中P16蛋白表达的意义

目的研究原发性肝细胞癌中Ｐ１６蛋白的表达程度与转移及预后的关系．方法应用免疫组织化学ＬＳＡＢ法对４５例原发性肝细胞癌组织及癌旁组织进行Ｐ１６蛋白表达的检测，并结合临床资料进行分析．结果肝癌肿瘤区及癌旁区Ｐ１６蛋白阳性表达率分别为４００％和８４４％，两组比较，有显著性差异（Ｐ＜００５）．Ｐ１６蛋白阳性表达率在无转移的肝癌中为６０５％，有转移组为２００％；高、中和低分化的肝癌分别为５００％，２７８％和２２２％．故Ｐ１６蛋白阳性表达在肝癌患者高分化与中和低分化比较有显著性差异（Ｐ＜００５），转移与非转移组比较有极显著性差异（Ｐ＜００１）．结论Ｐ１６蛋白表达与肝细胞癌增殖、分化程度及预后有关     

应用SELDI-TOF-MS技术筛选肺鳞癌和肺腺癌患者差异蛋白的研究

目的 应用表面增强激光解析电离飞行时间质谱(SELDI-TOF-MS)技术筛选出不同病理类型肺癌患者血清、支气管肺泡灌洗液(BALF)和肺癌组织中的差异蛋白,并探讨其临床意义.方法 选用WCX-2芯片、SELDI-TOF-MS技术检测20例肺鳞癌、20例肺腺癌和20例肺良性病变患者血清、BALF和肺组织匀浆的蛋白质谱,用Biomarker Pattern软件分析肺鳞癌和肺腺癌的差异蛋白并初步建立诊断模型.结果 ①肺鳞癌组与肺良性病变组:在血清、BALF、肺组织匀浆中发现存在2、9、8个差异蛋白峰(P＜0.05),分别选用其中质荷比为5 124.24、7 967.29联合10 843.45、7914.59联合8709.66的差异蛋白波峰建立分类树模型,其诊断肺鳞癌的灵敏度分别为85％、80％、75％,特异度分别为65％、80％、90％,正确率分别为75％、80％、83％,阳性预测值分别为71％、80％、88％,阴性预测值分别为81％、80％、78％,其相对应ROC曲线下面积分别为0.750、0.916、0.930.②肺腺癌组与肺良性病变组:在血清、BALF、肺组织匀浆中发现存在8、9、7个差异蛋白峰(P＜0.05),分别选用其中质荷比为9295.79、7923.01、2452.49的差异蛋白波峰建立分类树模型,其诊断肺腺癌的灵敏度分别为75％、80％、70％,特异度分别为65％、90％、85％,正确率分别为70％、85％、78％,阳性预测值分别为68％、89％,82％,阴性预测值分别为72％、82％、71％,其相对应ROC曲线下面积分别为0.844、0.933、0.825.结论 肺鳞癌和肺腺癌BALF及肺组织匀浆中的差异蛋白较血清多,诊断效率比血清高,肺鳞癌BALF中质荷比为7967.29和10843.45的差异蛋白峰联合、肺组织匀浆中质荷比为7914.59和8709.66的差异蛋白峰联合建模及肺腺癌BALF中质荷比为7923.01的差异蛋白峰建立分类树诊断模型,其诊断的灵敏度、特异度和正确率可达到75％～90％,有可能作为诊断肺鳞癌和肺腺癌的标志蛋白.     

DNA-PKcs蛋白和MGMT蛋白在肝癌组织中的表达及意义

目的探讨DNA依赖蛋白激酶催化亚单位（DNA—PKcs）和O^6-甲基鸟嘌呤-DNA甲基转移酶（MGMT）在肝癌发生、发展中的作用及临床应用价值。方法应用免疫组化方法检测DNA—PKcs蛋白和MGMT蛋白在68份肝癌组织（肝癌组）和10份正常肝脏组织（对照组）中的表达情况,采用X^2检验及Spearman秩和相关检验分析结果。结果肝癌组及对照组DNA—PKcs蛋白表达阳性率分别为85％和10％,MGMT蛋白表达阳性率分别为40％和90％（P均〈0．05）;两者表达具有负相关性（r=-0．475,P〈0．05）。结论DNA-PKcs和MGMT在肝癌发生、发展中起重要作用：有望作为肿瘤标记物、基因靶点等用于肝癌的诊治。     

WCX磁珠联合MALDI-TOF-MS技术在结直肠癌诊断中的应用

目的:比较结直肠癌与正常人血清蛋白质谱的变化,筛选特异性蛋白标志物,建立结直肠癌诊断分类树模型.方法:收集血清样本133例(其中结直肠癌67例,正常人66例),随机分为建模组和验证组.运用弱阳离子纳米磁珠(magnetic bead-weak cation exchange,MB-WCX)联合基质辅助激光解吸离子飞行质谱(matrix-assistedlaser desorption/ionization time-of-flight massspectrometry,MALDI-TOF-MS),建立结直肠癌与正常人血清蛋白质谱.用Flex Analysis2.4软件收集数据,应用ClinproTools2.2软件对建模组34例结直肠癌和33例正常人血清差异蛋白质谱进行定量分析,应用Genetic Algorithm算法建立结直肠癌诊断模型,应用所获取的诊断模型对验证组样本(33例结直肠癌和33例正常人)进行分类诊断,以评价诊断模型的诊断价值.结果:通过比较分析结直肠癌与正常人血清蛋白质谱,发现共有33个差异蛋白峰(P<0.05),其中在结直肠癌中表达上调25个,表达下调8个.利用其中5个差异峰(Mr分别为759,3316,4645,4248,2645Dr)建立诊断模型,获得了94.12%(32/34)敏感性和96.97%(32/33)的特异性,经独立样本双盲验证,其灵敏度为93.94%(31/33),特异度为96.97%(32/33).结论:基于磁珠分离和MALDI-TOF-MS技术能直接检测出结直肠癌患者血清差异表达蛋白,建立的诊断模型具有较高的敏感性和特异性,对提高结直肠癌的诊断具有一定的临床意义.     

Nomogram for predicting distant metastasis of male breast cancer: A SEER population-based study

The main purpose of this study was to build a prediction model for male breast cancer (MBC) patients to predict the possibility of distant metastasis. The Surveillance, Epidemiology, and End Results database was used to obtain data on patients with MBC. The patients were divided into a training set and a validation set at a ratio of 7:3. The risk variables of distant metastasis in the training set were determined by univariate and multivariate logistic regression analyses. And then we integrated those risk factors to construct the nomogram. The prediction nomogram was further verified in the verification set. The discrimination and calibration of the nomogram were evaluated by the area under the receiver operating characteristic curve, calibration plots, respectively. A total of 1974 patients (1381 in training set and 593 in validation set) were eligible for final inclusion, of whom 149 (7.55%) had distant metastasis at the diagnosed time. Multivariate logistic regression analyses presented that age, T stage, N stage, and hormone receptor status were independent risk factors for distant metastasis at initial diagnosis of male breast cancer. Finally, the 4 variables were combined to construct the nomogram. The area under the curve values for the nomogram established in the training set and validation set were 0.8224 (95%CI: 0.7796–0.8652) and 0.8631 (95%CI: 0.7937–0.9326), suggesting that the nomogram had good predictive power. The calibration plots illustrated an acceptable correlation between the prediction by nomogram and the actual observation, as the calibration curve was closed to the diagonal bisector line. An easy-to-use nomogram, being proven to be with reliable discrimination ability and accuracy, was established to predict distant metastasis for male patients with breast cancer using the easily available risk factors.     

The clinical benefits of performing staging laparoscopy for pancreatic cancer treatment

BackgroundThe indications and benefits derived from staging laparoscopy (SL) for pancreatic cancer (PC) remain controversial.MethodsThis study involved PC patients in whom resection had been considered possible between 2009 and 2020. We classified the patients into before 2014 (training set) and 2014 and later (validation set) groups, as SL was introduced in 2014, in our institution. In the training set, the predictors of non-curative factors were investigated, and reproducibility was confirmed in the validation set. In addition, the outcomes were compared between the datasets.ResultsA total of 802 patients were classified into the training set (n = 241) and validation set (n = 561). In the training set, pancreatic body or tail tumors (odds ratio [OR]: 2.62: P = 0.039), CA19-9 > 88 U/ml (OR: 3.21: P = 0.018) and a tumor diameter >36 mm (OR: 6.07; P < 0.001) were independent predictors of non-curative factors. The increased rate of non-curative factors was confirmed as the number of predictors increased in the validation set. The curative resection (CR) rate was significantly higher in the validation set than in the training set (P = 0.035). Although there was no significant difference in the OS in the not-resected group (P = 0.895), the OS in the CR and non-CR group was significantly better in the validation set than in the training set (CR, P < 0.001; non-CR, P < 0.001).ConclusionThe findings suggest potential candidates for SL and revealed improved outcomes by the advent of treatment strategies including SL.     

A prediction model for 30-day mortality of sepsis patients based on intravenous fluids and electrolytes

To establish a prediction model for the 30-day mortality in sepsis patients. The data of 1185 sepsis patients were extracted from the Medical Information Mart for Intensive Care III (MIMIC-III) and all participants were randomly divided into the training set (n = 829) and the testing set (n = 356). The model was established in the training set and verified in the testing set. After standardization of the data, age, gender, input, output, and variables with statistical difference between the survival group and the death group in the training set were involved in the extreme gradient boosting (XGBoost) model. Subgroup analysis was performed concerning age and gender in the testing set. In the XGBoost model with variables related to intravenous (IV) fluid management and electrolytes for the 30-day mortality of sepsis patients, the area under the curve (AUC) was 0.868 (95% confidence interval [CI]: 0.867–0.869) in the training set and 0.781 (95% CI: 0.779–0.782) in the testing set. The sensitivity was 0.815 (95% CI: 0.774–0.857) in the training set and 0.755 (95% CI: 0.686–0.825) in the testing set. The specificity was 0.761 (95% CI: 0.723–0.798) in the training set, and 0.737 (95% CI: 0.677–0.797) in the testing set. In the XGBoost forest model without variables related to IV fluid management and electrolytes for the 30-day mortality of sepsis patients, in the training set, the AUC was 0.830 (95% CI: 0.829–0.831), the sensitivity was 0.717 (95% CI: 0.669–0.765), the specificity was 0.797 (95% CI: 0.762–0.833), and the accuracy was 0.765 (95% CI: 0.736–0.794). In the testing set, the AUC was 0.751 (95% CI: 0.750–0.753), the sensitivity was 0.612 (95% CI: 0.533–0.691), the specificity was 0.756 (95% CI: 0.698–0.814), and the accuracy was 0.697(95% CI: 0.649–0.744). The prediction model including variables associated with IV fluids and electrolytes had good predictive value for the 30-day mortality of sepsis patients.     

Interobserver agreement for EUS findings in familial pancreatic-cancer kindreds

BACKGROUND: EUS is a promising modality for pancreatic-cancer screening in high-risk persons, including familial pancreatic-cancer (FPC) kindreds. OBJECTIVE: To assess interobserver agreement for interpretation of EUS in persons at high risk for pancreatic cancer. DESIGN: Seventeen expert endosonographers blinded to patients' clinical history rated a "training set" of pancreatic EUS video clips for the presence of a normal examination, masses, cysts, and features of chronic pancreatitis. Clips included high-risk persons and controls (normal and various pancreatic diseases). The endosonographers then participated in a workshop on EUS findings in high-risk persons and drafted a consensus statement. Three months later, they blindly rated a "test set" composed of the same video clips. MAIN OUTCOME MEASUREMENTS: Interobserver agreement at baseline (training set) and after a consensus process (test set). RESULTS: For the training set, interobserver agreement was good (kappa>or=0.4) for the presence of cysts and was fair to poor for all other rated EUS features and diagnosis of normal. There was no overall improvement in the test set. In both the training and test sets, agreement was worse for clips from FPC kindreds (kappa>or=0.4 for cysts and <0.4 for all other features) than for controls (kappa>or=0.4 for normal, cysts, masses, echogenic strands, and lobularity). LIMITATIONS: Video clips were not of identical image quality and duration as a clinical EUS examination. CONCLUSIONS: There was fair to poor interobserver agreement for the interpretation of pancreatic EUS video clips from members of FPC kindreds. Agreement was not improved by a consensus process.     

The effect of different intensity modalities of resistance training on beat-to-beat blood pressure in cardiac patients.

BACKGROUND: Resistance training has been introduced in cardiac rehabilitation to give more benefit than traditional training. Haemodynamic evaluation of cardiac patients to resistance training has generally consisted of continuous HR monitoring and discontinuous blood pressure measurements. DESIGN AND METHODS: Blood pressure (BP) and heart rate (HR) responses to resistance training were evaluated using continuous monitoring (Finapres) during low (four sets of 17 repetitions at 40% of the one-repetition maximum strength [1-RM]) and high intensity resistance training (four sets of 10 repetitions at 70% of 1-RM) on a leg extension machine in 14 patients who participated in a rehabilitation programme. Work volume was identical in the low- and high-level resistance training. RESULTS: The HR and systolic blood pressure (SBP) during low intensity resistance training were always larger than during high intensity (P<0.001). Peak SBP increased from set 1 to set 3 and 4 during both low and high intensity resistance training (P<0.05). Peak HR was larger in set 4 (95+/-11 bpm) than in set 1 only during low intensity resistance training (91+/-12 bpm) (P<0.05). One-minute recovery periods did not allow a return to baseline HR and SBP during both low and high intensity modalities. CONCLUSIONS: The SBP and HR responses to resistance training are related to the duration of exercise. Sets with < or =10 repetitions of high intensity should be preferred to longer sets with low intensity. Pauses between exercise sets should exceed 1 min. Blood pressure should be measured during the last repetitions of the exercise set.     

Application of the HALF index obviates the need for liver biopsy in half of all patients with chronic hepatitis B

Background and Aims: Transient elastography (TE) is useful for predicting the fibrosis stage, but it is unsatisfactory as a substitute for liver biopsy, especially in patients with chronic hepatitis B (CHB). This study was performed to establish a reliable model for predicting significant fibrosis (SF) in patients with CHB. Methods: All CHB patients who were admitted to undergo liver biopsy were enrolled. They were randomly classified into either a training set (n = 139) or a validation set (n = 69). A model for predicting SF was established in the training set and validated in the validation set. Low and high cutoff values (COVs) were chosen for sensitivity ≥ 99% and specificity ≥ 99%, respectively. Results: A total of 208 patients were enrolled. Age was 39 ± 12 years and 149 (71.6%) were men. In the training set, liver stiffness values and serum haptoglobin, apolipoprotein A1, and α2‐macroglobulin levels were independent predictors of SF on multivariate analysis. These variables were used to construct a novel model, called the HALF index. The area under the receiver operating characteristics curve of the HALF index for predicting SF was significantly higher than that of TE alone (0.915 vs 0.877, P = 0.010). Using low and high COVs of the HALF index, it appears that approximately half (47.1%) of patients could avoid liver biopsy, with an associated accuracy of 99.0%. Conclusion: A combination of liver stiffness and serum markers identified SF with a high degree of accuracy. Approximately half of all patients with CHB could avoid liver biopsy through the utilization of the HALF index.     

A novel gas chromatography mass spectrometry-based serum diagnostic and assessment approach to ulcerative colitis

Background & aimsTo improve the clinical course of ulcerative colitis (UC), more accurate serum diagnostic and assessment methods are required. We used serum metabolomics to develop diagnostic and assessment methods for UC.MethodsSera from UC patients, Crohn's disease (CD) patients, and healthy volunteers (HV) were collected at multiple institutions. The UC and HV were randomly allocated to the training or validation set, and their serum metabolites were analyzed by gas chromatography mass spectrometry (GC/MS). Using the training set, diagnostic and assessment models for UC were established by multiple logistic regression analysis. Then, the models were assessed using the validation set. Additionally, to establish a diagnostic model for discriminating UC from CD, the CD patients' data were used.ResultsThe diagnostic model for discriminating UC from HV demonstrated an AUC of 0.988, 93.33% sensitivity, and 95.00% specificity in the training set and 95.00% sensitivity and 98.33% specificity in the validation set. Another model for discriminating UC from CD exhibited an AUC of 0.965, 85.00% sensitivity, and 97.44% specificity in the training set and 83.33% sensitivity in the validation set. The model for assessing UC showed an AUC of 0.967, 84.62% sensitivity, and 88.23% specificity in the training set and 84.62% sensitivity, 91.18% specificity, and a significant correlation with the clinical activity index (r_s = 0.7371, P < 0.0001) in the validation set.ConclusionsOur models demonstrated high performance and might lead to the development of a novel treatment selection method based on UC condition.     

用蛋白质芯片技术筛选非小细胞肺癌患者血清中标志蛋白

目的 探讨用蛋白质芯片技术检测血清非小细胞肺癌（NSCLC）标志蛋白筛查肺癌患者的可行性。方法 用蛋白质芯片表面增强激光解吸电离飞行时间质谱仪（SELDI-TOF-MS）技术检测123例肺癌患者和40名正常人血清蛋白质质谱。用数字表法随机抽取94份标本（53例NSCLC，21例小细胞肺癌和20名正常人）作为训练组进行系统训练，将筛选出来的相对分子质量为11493、6429、8245、5336及2536的5个蛋白峰作为一个标志物组合模式，建立分类树模型（即系统训练过程）；用69份未知血清标本（49例NSCLC，20名正常人）作为盲筛组验证该模型。结果 系统显示，在训练组该模式检测NCLC的敏感性和特异性分别为95．9％（71／74）、90．0％（18／20），盲筛组分别为83．7％（41／49）及80．0％（16／20）。结论 蛋白质芯片SELDI-TOF-MS技术能较准确的区分NSCLC患者与健康对照者，该技术为NSCLC的筛查提供了新的有效工具。     

Neural networks for classification of ECG ST-T segments.

The usefulness of neural networks for pattern recognition in electrocardiographic (ECG) ST-T segments was assessed. Two thousand ST-T segments from the 12-lead ECG were visually classified singly into 7 different groups. The material was divided into a training set and a test set. Computer-measured ST-T data for each element in the training set, paired with the corresponding classification, was input to various configurations of software-based neural networks during a learning process. Thereafter, the networks correctly classified 90-95% of the individual ST-T segments in the test set. The importance of the size and composition of the training set in determining the performance of a network was clearly demonstrated. In conclusion, neural networks can be used for classification of ST-T segments. If carefully incorporated into a conventional ECG interpretation program, neural networks may well be of value for automated ECG interpretation in the near future.     

Pattern recognition technique in immunological antigenic tests to identify Mycobacterium tuberculosis infection

The importance of a diagnostic test that is simple and quick to identify Mycobacterium tuberculosis infection needs no emphasis. The tuberculin skin test (TST - 1 TU RT23) is the diagnostic tool for identifying M. tuberculosis infection at present. The test reaction on the skin is measured after 48-72 h. It is observed that often multi-modes are seen, when the reactions are drawn as a graph and the bimodality is seen very feebly. Because of the difficulties in the administration of TST, several serological tests were developed over three decades, but none of the studies showed the desired results. One study to evaluate purified protein derivative (PPD) antigen resulted in a claim of 80% sensitivity and 4% false-positivity rate (14), while other researchers were not able to obtain similar results. In addition, several problems were encountered due to the non-availability of antigens, and data analyses from an ELISA-based diagnostic test showed considerable overlap of distributions of optical density (OD) values among patients and healthy individuals (10). Classical statistical techniques cannot explain the cause of these overlaps. Hence, an attempt is made in this article to resolve these difficulties by the pattern recognition technique (PRT). The technique lies in splitting the data into clusters using a supervised algorithm. The data set is normally split into a training set, a test set and a validation set. The PRT gets "trained" through the training data set until the infected and uninfected groups of individuals are correctly classified. The training occurs based on an algorithm on the training set. On successful completion of the training, this technique is further tested and validated in the respective data sets. SETTING: A total of 273 finger-prick specimens were collected from five categories (Al, A2, B, C, D, E) of subjects not vaccinated with (bacille Calmette Guerin) (BCG) from Trivellore BCG Trial area adopted by the Tuberculosis Research Center, Chennai, India. OBJECTIVE: The study was conducted with the primary aim of evaluating purified antigens--r38 kDa, PPD and 30 kDa--for their usefulness as diagnostic tools and to test the applicability of the PRT in the evaluation of diagnostic tests. Individuals in two main categories (definitely not infected categories Al, A2 and D, and definitely infected categories B, E and C based on reaction to TST) were assembled for the purpose. RESULTS: The overall PRT performance of 30 kDa was 72.3% sensitivity and 90.9% specificity for identifying M. tuberculosis infection, while the r38 kDa antigen recorded a sensitivity of 73.8% and a specificity of 84.6%. In the case of PPD, the results were not promising. CONCLUSION: This paper on ELISA-based diagnostic tests attempts to implement an optimal decision support system through PRT that would identify the outcome (as infected or non-infected) based on the OD values. The PRT was able to predict the outcome for individual suspects. Further, Kullback-Leibler distance measurement has validated the PRT in distinguishing infected individuals from healthy subjects (based on the OD values).