Automated Computer Algorithms Research Articles

Automated analysis of postings in fourth grade online discussions to help teachers keep students on-track

ABSTRACT In the recent, and ongoing, Covid-19 pandemic, remote or online K-12 schooling became the norm. Even if the pandemic tails off somewhat, remote K-12 schooling will likely remain more frequent than it was before the pandemic. A mainstay technique of online learning, at least at the college and graduate level, has been the online discussion. Since it does afford the potential for meaningful learner-learner and instructor-learner interaction, which are vital for distance learning, it is worth considering online discussions for K-12 remote schooling. One challenge with online learning in general, and online discussion in particular, is that it is labor intensive for teachers to moderate. Effective moderating of online discussions is vital for discussions to be nurturing, effective learning situations. Yet, moderating of online discussions is notoriously labor-intensive for teachers/instructors. Further, since younger learners are more likely to drift off topic, in general, but particularly in small group online discussions, automated early warning systems are helpful. The current study collected small group, “book club”, discussion data from fourth graders reading web-based eBooks in Slovenian primary schools, qualitatively coded the data and analyzed postings using computer-based natural language processing to predict when students went off-topic. One indicator that postings are on-topic is book relevance, i.e. that the posting is relevant to eBook content. The computer algorithm correctly predicted book relevance of postings 90 percent of the time, suggesting that automated computer algorithms could assist teachers with moderating online discussions, providing real-time notifications of problems in online discussions. Further, this study provided a proof-of-concept that small group online discussions, in web-based eBooks can be practical and educationally meaningful in fourth grade classes.

Developing Automated Computer Algorithms to Track Periodontal Disease Change from Longitudinal Electronic Dental Records

To develop two automated computer algorithms to extract information from clinical notes, and to generate three cohorts of patients (disease improvement, disease progression, and no disease change) to track periodontal disease (PD) change over time using longitudinal electronic dental records (EDR). We conducted a retrospective study of 28,908 patients who received a comprehensive oral evaluation between 1 January 2009, and 31 December 2014, at Indiana University School of Dentistry (IUSD) clinics. We utilized various Python libraries, such as Pandas, TensorFlow, and PyTorch, and a natural language tool kit to develop and test computer algorithms. We tested the performance through a manual review process by generating a confusion matrix. We calculated precision, recall, sensitivity, specificity, and accuracy to evaluate the performances of the algorithms. Finally, we evaluated the density of longitudinal EDR data for the following follow-up times: (1) None; (2) Up to 5 years; (3) > 5 and ≤ 10 years; and (4) >10 and ≤ 15 years. Thirty-four percent (n = 9954) of the study cohort had up to five years of follow-up visits, with an average of 2.78 visits with periodontal charting information. For clinician-documented diagnoses from clinical notes, 42% of patients (n = 5562) had at least two PD diagnoses to determine their disease change. In this cohort, with clinician-documented diagnoses, 72% percent of patients (n = 3919) did not have a disease status change between their first and last visits, 669 (13%) patients' disease status progressed, and 589 (11%) patients' disease improved. This study demonstrated the feasibility of utilizing longitudinal EDR data to track disease changes over 15 years during the observation study period. We provided detailed steps and computer algorithms to clean and preprocess the EDR data and generated three cohorts of patients. This information can now be utilized for studying clinical courses using artificial intelligence and machine learning methods.

Developing Automated Computer Algorithms to Phenotype Periodontal Disease Diagnoses in Electronic Dental Records.

Our objective was to phenotype periodontal disease (PD) diagnoses from three different sections (diagnosis codes, clinical notes, and periodontal charting) of the electronic dental records (EDR) by developing two automated computer algorithms. We conducted a retrospective study using EDR data of patients (n = 27,138) who received care at Temple University Maurice H. Kornberg School of Dentistry from January 1, 2017 to August 31, 2021. We determined the completeness of patient demographics, periodontal charting, and PD diagnoses information in the EDR. Next, we developed two automated computer algorithms to automatically diagnose patients' PD statuses from clinical notes and periodontal charting data. Last, we phenotyped PD diagnoses using automated computer algorithms and reported the improved completeness of diagnosis. The completeness of PD diagnosis from the EDR was as follows: periodontal diagnosis codes 36% (n = 9,834), diagnoses in clinical notes 18% (n = 4,867), and charting information 80% (n = 21,710). After phenotyping, the completeness of PD diagnoses improved to 100%. Eleven percent of patients had healthy periodontium, 43% were with gingivitis, 3% with stage I, 36% with stage II, and 7% with stage III/IV periodontitis. We successfully developed, tested, and deployed two automated algorithms on big EDR datasets to improve the completeness of PD diagnoses. After phenotyping, EDR provided 100% completeness of PD diagnoses of 27,138 unique patients for research purposes. This approach is recommended for use in other large databases for the evaluation of their EDR data quality and for phenotyping PD diagnoses and other relevant variables.

The Impact of Automated Information Acquisition on the Stock Market

Does collecting regulatory financial information with the help of automated computer algorithms (robots) affect the stock market? Using the EDGAR Server Log data set, I construct firm-level measures of information acquisition by robots and non-robots and show that robots are extensively used for information acquisition when new information becomes available. The SEC's mandate regarding interactive data leads to a notable increase in information demand, consistent with decreased information acquisition costs for standardised regulatory financial information in XBRL-format. A higher relative importance of robots acquiring information about a firm combined with the XBRL adoption is associated with a consequent increase in trading volume, smaller bid-ask spreads, lower volatility, positive cumulative abnormal return and increased volume coefficient of variation. The findings are consistent with the idea that automation and standardisation benefits informed investors disproportionately more than uninformed traders.

Patient-Ventilator Interaction During Noninvasive Ventilation in Subjects With Exacerbation of COPD: Effect of Support Level and Ventilator Mode.

Patient-ventilator synchrony in patients with COPD is at risk during noninvasive ventilation (NIV). NIV in neurally-adjusted ventilatory assist (NAVA) mode improves synchrony compared to pressure support ventilation (PSV). The current study investigated patient-ventilator interaction at 2 levels of NAVA and PSV mode in subjects with COPD exacerbation. NIV was randomly applied at 2 levels (5 and 15 cm H2O) of PSV and NAVA. Patient-ventilator interaction was evaluated by comparing airway pressure and electrical activity of the diaphragm waveforms with automated computer algorithms. 8 subjects were included. Trigger delay was longer in PSV high (268 ± 112 ms) than in PSV low (161 ± 118 ms, P = .043), and trigger delay during NAVA was shorter than PSV for both low support (49 ± 24 ms for NAVA, P = .035) and high support (79 ± 276 ms for NAVA, P = .003). No difference in cycling error for low and high levels of PSV (PSV low -100 ± 114 ms and PSV high 56 ± 315 ms) or NAVA (NAVA low -5 ± 18 ms, NAVA high 12 ± 36 ms) and no difference between PSV and NAVA was found. Increasing PSV levels during NIV caused a progressive mismatch between neural effort and pneumatic timing. Patient-ventilator interaction during NAVA was more synchronous than during PSV, independent of inspiratory support level. (ClinicalTrials.gov registration NCT01791335.).

Automated image quality assessment for chest CT scans.

Medical image quality needs to be maintained at standards sufficient for effective clinical reading. Automated computer analytic methods may be applied to medical images for quality assessment. For chest CT scans in a lung cancer screening context, an automated quality assessment method is presented that characterizes image noise and image intensity calibration. This is achieved by image measurements in three automatically segmented homogeneous regions of the scan: external air, trachea lumen air, and descending aorta blood. Profiles of CT scanner behavior are also computed. The method has been evaluated on both phantom and real low-dose chest CT scans and results show that repeatable noise and calibration measures may be realized by automated computer algorithms. Noise and calibration profiles show relevant differences between different scanners and protocols. Automated image quality assessment may be useful for quality control for lung cancer screening and may enable performance improvements to automated computer analysis methods.

Breast MRI contrast enhancement kinetics of normal parenchyma correlate with presence of breast cancer.

BackgroundWe investigated dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) contrast enhancement kinetic variables quantified from normal breast parenchyma for association with presence of breast cancer, in a case-control study.MethodsUnder a Health Insurance Portability and Accountability Act compliant and Institutional Review Board-approved protocol, DCE-MRI scans of the contralateral breasts of 51 patients with cancer and 51 controls (matched by age and year of MRI) with biopsy-proven benign lesions were retrospectively analyzed. Applying fully automated computer algorithms on pre-contrast and multiple post-contrast MR sequences, two contrast enhancement kinetic variables, wash-in slope and signal enhancement ratio, were quantified from normal parenchyma of the contralateral breasts of both patients with cancer and controls. Conditional logistic regression was employed to assess association between these two measures and presence of breast cancer, with adjustment for other imaging factors including mammographic breast density and MRI background parenchymal enhancement (BPE). The area under the receiver operating characteristic curve (AUC) was used to assess the ability of the kinetic measures to distinguish patients with cancer from controls.ResultsWhen both kinetic measures were included in conditional logistic regression analysis, the odds ratio for breast cancer was 1.7 (95 % CI 1.1, 2.8; p = 0.017) for wash-in slope variance and 3.5 (95 % CI 1.2, 9.9; p = 0.019) for signal enhancement ratio volume, respectively. These odds ratios were similar on respective univariate analysis, and remained significant after adjustment for menopausal status, family history, and mammographic density. While percent BPE was associated with an odds ratio of 3.1 (95 % CI 1.2, 7.9; p = 0.018), in multivariable analysis of the three measures, percent BPE was non-significant (p = 0.897) and the two kinetics measures remained significant. For the differentiation of patients with cancer and controls, the unadjusted AUC was 0.71 using a combination of the two measures, which significantly (p = 0.005) outperformed either measure alone (AUC = 0.65 for wash-in slope variance and 0.63 for signal enhancement ratio volume).ConclusionsKinetic measures of wash-in slope and signal enhancement ratio quantified from normal parenchyma in DCE-MRI are jointly associated with presence of breast cancer, even after adjustment for mammographic density and BPE.

Abstract P1-01-07: Quantitative assessment of early- and delayed DCE-MRI background parenchymal enhancement in breast cancer risk prediction

Abstract Purpose Background parenchymal enhancement (BPE) estimated from breast dynamic contrast enhanced MRI (DCE-MRI) has been correlated with breast cancer risk. Multiple time-point post-contrast-subtracted sequences are acquired in typical clinical breast DCE-MRI protocols. We quantitatively assessed BPE using fully automated computer algorithms and investigated the relationship between BPE quantified from three time-point post-contrast sequences and breast cancer risk prediction. Materials and Methods A retrospective case-control study was performed using breast DCE-MRI scans from 102 patients (mean 47.2±7.3 YO) who underwent either open surgical biopsy or core biopsy from 2009-2011: 51 women had unilateral breast cancer and 51 were age- and date-of-MRI matched controls with a unilateral biopsy-proven benign. For each MRI scan, three post-contrast-subtracted sequences (i.e., SUB 1, SUB 2 and SUB 3) acquired over a total of 7 minutes were analyzed. BPE was quantified from each of the sequences using fully automated computer algorithms on the breasts contralateral to the cancers and the contralateral (negative) breasts of the controls. For each sequence, two quantitative BPE measures were generated: the absolute BPE volume (|BPE|) and its relative amount over the whole breast volume (BPE%). Volumetric absolute and relative amounts of fibroglandular tissue (|FGT| and FGT%) were also automatically quantified, from the pre-contrast sequence. BI-RADS breast density assessment was retrieved from the mammography report (< 6 months) prior to cancer diagnosis (for cases) or MRI (for controls). Multivariable conditional logistic regression was performed to assess BPE measures, quantified respectively from SUB 1, SUB 2, and SUB 3, as predictors of breast cancer risk. Results Breast cancer risk odds ratios (ORs) were reported by |BPE| and BPE% (Table 1), after adjustment for breast density, |FGT|, and FGT%. Breast density did not correlate with risk for breast cancer in this study cohort: OR for BI-RADS density alone was 0.75 (95% CI: 0.35, 1.59; p=0.5). OR was 1.14 (95% CI: 0. 72, 1.81; p=0.6) for |FGT| alone and 0.70 (95% CI: 0.19, 2.52; p=0.6) for FGT% alone. Table 1: Breast cancer risk odds ratios (ORs) with 95% confidence intervals [CIs] for BPE quantified from SUB 1, SUB 2, and SUB 3, respectively. BPE from SUB1BPE from SUB2BPE from SUB3|BPE|2.0 (95% CI: 1.1-3.7; p=0.02)2.0 (95% CI: 1.2-3.3; p=0.01)1.7 (95% CI: 1.1-2.7; p=0.02)BPE%3.6 (95% CI: 1.3-10.1; p=0.01)2.8 (95% CI: 1.3-6.2; p=0.01)2.4 (95% CI: 1.2-5.0; p=0.02) Conclusions Increased BPE quantified from DCE-MRI are predictive of breast cancer risk, independent of measures of breast density and FGT. BPE estimated from three time-point SUB sequences has a similar predictive effect of breast cancer risk, while an early sequence (e.g., SUB 1) appears to have a larger magnitude of effect than that of a delayed sequence (e.g., SUB 3). Clinical Relevance Quantified BPE in breast DCE-MRI has potential for use as a biomarker of breast cancer risk and may be included to improve breast cancer risk prediction. A single post-contrast sequence (i.e., SUB 1) may be adequate for use to estimate breast cancer risk using breast MRI in the context of breast cancer screening for high-risk women. Citation Format: Shandong Wu, Wendie A Berg, Margarita L Zuley, Brenda F Kurland, Rachel C Jankowitz, Jules Sumkin, Robert M Nishikawa, David Gur. Quantitative assessment of early- and delayed DCE-MRI background parenchymal enhancement in breast cancer risk prediction [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P1-01-07.

Abstract P1-01-06: Quantitative breast DCE-MRI risk biomarkers and breast tumor immunohistochemistry phenotypes: A preliminary assessment

Abstract Purpose Fibroglandular tissue (FGT) and background parenchymal enhancement (BPE) estimated from breast dynamic contrast enhanced MRI (DCE-MRI) have been correlated with breast cancer risk. We performed a preliminary study assessing in a small cohort of breast cancer patients the relationship between FGT and BPE both quantified from cancer-unaffected breasts and breast tumor’s immunohistochemistry phenotypes, including testing of estrogen (ER), progesterone (PR), and human epidermal growth factor receptor 2 (HER-2/neu) receptors. Materials and Methods We retrospectively identified 51 breast cancer patients who were diagnosed with unilateral breast cancer from 2009-2011. Six out of the 51 were excluded for analysis due to missing data in ER, PR, or Her-2 testing in their pathology reports. The 45 cancer patients (mean 47.3±7.6 YO; 24 IDC, 4 ILC, 16 mixed of IDC and DCIS, and 1 papillary carcinoma) consist of 25 premenopausal and 20 postmenopausal women. Breast DCE-MRI scans acquired at time-of-diagnosis were analyzed using fully automated computer algorithms. From the contralateral cancer-free breast, four quantitative variables were computed for each patient: the absolute volumes of FGT and BPE (i.e., |FGT| and |BPE|) and their relative amount over the whole-breast volume (i.e., FGT% and BPE%). BPE were assessed over the FGT region of the breast, and separately from each of the three sequential DCE post-contrast-subtracted sequences (i.e., SUB1, SUB2, and SUB3). Multivariable logistic regressions were performed to assess the quantitative FGT/BPE variables for distinguishing/predicting the status (+/-) of ER, PR, and HER-2 testing. Area under the curve (AUC) of the receiver operating characteristic (ROC) was used to evaluate the regression models’ discriminative performance. Odds ratios (ORs) were reported with 95% confidence intervals (CIs). Results In this cohort, the most common phenotypes were ER+/PR+ (14/45 each HER2+ and HER2-) and triple negative (9/45). Premenopausal status was associated with tumors that were HER-2+ (AUC=0.66; p=0.04), ER+ (AUC=0.73; p=0.02), and PR+ (ACU=0.66; p=0.04). FGT% (AUC=0.74; p=0.017; OR=5.3 [95% CI: 1.3, 21.1]) and BPE%-SUB3 (AUC=0.69; p=0.02; OR = 1.4 [95%CI: 1.1, 1.9]) were discriminative of HER-2 status; after adjusting for menopausal status, FGT% remained the association (AUC=0.77; p=0.04; OR=4.4 [95% CI: 1.1, 18.5]) but BPE%-SUB3 did not (p=0.06). BPE%-SUB1 was discriminative of PR status (AUC=0.70; p=0.03; OR=2.3 [95%CI: 1.1, 4.8]) but did not contribute to predict when adjusting for menopausal status (AUC=0.75; p=0.06). All FGT and BPE variables were not associated (p>0.05) with ER status. Conclusions In this preliminary study, FGT% and BPE% were associated with HER-2 and PR status, and none of FGT and BPE was associated with ER status, of breast tumor. Clinical Relevance Quantitative MRI variables of FGT and BPE that are predictive of breast cancer risk are related to breast tumor immunohistochemistry phenotypes. While FGT and BPE cannot replace the testing of these phenotypes, they may contribute to assess the risk of molecular subtypes of breast cancer, with potential implications for improving breast cancer screening strategies. Citation Format: Shandong Wu, Margarita L Zuley, Brenda F Kurland, Rachel C Jankowitz, Jules Sumkin, David Gur. Quantitative breast DCE-MRI risk biomarkers and breast tumor immunohistochemistry phenotypes: A preliminary assessment [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P1-01-06.

EFIS—Evolving Fuzzy Image Segmentation

Despite the large number of techniques proposed in recent years, accurate segmentation of digital images remains a challenging task for automated computer algorithms. Approaches based on machine learning hold particular promise in this regard, because in many applications, e.g., medical image analysis, frequent user intervention can be assumed to correct the results, thereby generating valuable feedback for algorithmic learning. In order to learn segmentation of new (unseen) images, such user feedback (correction of current or past results) seems indispensable. In this paper, we propose the formation and evolution of fuzzy rules for user-oriented environments in which feedback is captured by design. The evolving fuzzy image segmentation (EFIS) can be used to adjust the parameters of existing segmentation methods, switch between their results, or fuse their results. Specifically, we propose a single-parametric EFIS (SEFIS), apply its rule evolution to breast ultrasound images, and evaluate the results using three segmentation methods, namely, global thresholding, region growing, and statistical region merging. The results show increased accuracy across all tests and for all methods. For instance, the accuracy of statistical region merging can be improved from 59% ± 30% to 71% ± 22%. We also propose a multiparametric EFIS (MEFIS) for switching between or fusing the results of multiple segmentation methods. Preliminary results indicate that MEFIS can further increase overall segmentation accuracy. Three thresholding methods with accuracies of 62% ± 11%, 64% ± 16%, and 61% ± 9% were combined to reach an overall accuracy of 66% ± 15%. Finally, we compare our SEFIS scheme with five other thresholding methods to evaluate its overall performance.

Heterogeneity and matching of ventilation and perfusion within anatomical lung units in rats

Prior studies exploring the spatial distributions of ventilation and perfusion have partitioned the lung into discrete regions not constrained by anatomical boundaries and may blur regional differences in perfusion and ventilation. To characterize the anatomical heterogeneity of regional ventilation and perfusion, we administered fluorescent microspheres to mark regional ventilation and perfusion in five Sprague–Dawley rats and then using highly automated computer algorithms, partitioned the lungs into regions defined by anatomical structures identified in the images. The anatomical regions ranged in size from the near-acinar to the lobar level. Ventilation and perfusion were well correlated at the smallest anatomical level. Perfusion and ventilation heterogeneity were relatively less in rats compared to data previously published in larger animals. The more uniform distributions may be due to a smaller gravitational gradient and/or the fewer number of generations in the distribution trees before reaching the level of gas exchange, making regional matching of ventilation and perfusion less extensive in small animals.

Comparison of distance measures for manifold learning: Application to Alzheimer’s brain scans

The scale of medical imaging data is growing rapidly and automated computer algorithms are well suited to analyze such data. Shape information can distinguish diseased scans from normal controls, but analyzing the data is difficult due to the high dimensionality of shape information. With manifold learning, shape analysis becomes more tractable in a low dimensional space. Some manifold learning methods, including multidimensional scaling (MDS), require a distance measure to quantify pair-wise dissimilarities between scans of interest. In this study, we compared two different distance measures combined with MDS to distinguish patients with Alzheimer’s disease (AD) and mild cognitive impairment (MCI) from normal control patients. The first distance measure is based on the displacement field, and the second distance measure is based on mutual information (MI). Shape quantification was applied to the brain scans of 25 normal, 25 AD, and 25 MCI patients. Use of the first distance measure resulted in an 18% error rate while use of the second distance measure resulted in a 46% error rate for classifying between patients with AD and normal patients. Application of MDS leads to a feature space, and we compared the MDS-induced feature space with the feature space induced from hippocampus volume, a traditionally used feature for distinguishing AD/MCI patients from normal patients.

Elemental Analysis of Neutron-Induced Gamma-Ray Spectra for Detection of Hazardous Materials

Elemental analysis of neutron-induced gamma-ray spectra is a significant technology in the detection of chemical agents, explosives, etc. The hard part of this problem is the very complicated and uncertain background signals of the gamma-ray spectra. Also, the background signals are always changing as the searched objects change, thus further complicating the gamma-ray spectra analysis process. We can define a typical or average background spectrum if the variation of background spectrum is not too large, then we use this background spectrum to identify a gamma-ray signal.We tested both the direct summation approach and the Gaussian fitting approach in our computer algorithms. We found these two different approaches have individual advantages and disadvantages when they are applied in calculations of the signal significance level. In the end, we combined the direct summation approach with the Gaussian fitting approach in our computer algorithms in the actual calculations of the signal significance level.Based on our previous preliminary MCNP simulations results, we used phosphorus powder to simulate the chemical agent sarin and used our automated computer algorithms to calculate the single-line and multiple-line signal significance levels. Presented in this paper are some results in which we used our experimental data to test our average background spectrum and our computer algorithms. Our calculated results show the average background spectra that we defined are appropriate for the elemental analysis in searching for the chemical agent, and our computer algorithms, in which we combined the direct summation approach with the Gaussian fitting approach, function well. From a multiple-line analysis, our calculated results show that in the real application of this neutron-induced gamma-ray detection technique the detection time can be reduced to 15 s or less for detecting small quantities of chemical agents.

Automated sonographic evaluation of testicular perfusion

Contrast-enhanced ultrasound (US) imaging is potentially applicable to the investigation of vascular disorders of the testis. We investigated the ability of two automated computer algorithms to analyse contrast-enhanced pulse inversion US data in a rabbit model of unilateral testicular ischaemia and to correctly determine relative testicular perfusion: nonlinear curve fitting of the US backscatter intensity as a function of time; and spectral analysis of the intensity time trace. We compared (i) five metrics based on the algorithmic data to testicular perfusion ratios obtained with radiolabelled microspheres, a reference standard; (ii) qualitative assessment of the US images by two independent readers blinded to the side of the experimental and control testes to the radiolabelled microsphere perfusion ratios; and (iii) results of the algorithmically-derived metrics to the qualitative assessments of the two readers. For the curve fit method, the algorithmically-derived metrics agreed with the reference standard in 54% to 68% of all cases. For the spectral method, the results agreed in 70% of all cases. The two readers agreed with the reference standard in 40% and 35% of all cases, respectively. These results suggest that automated methods of analysis may provide useful information in the assessment of testicular perfusion.

Accrual of MRI white matter abnormalities in elderly with normal and impaired mobility

White matter signal abnormality (WMSA) is often present in the MRIs of older persons with mobility impairment. We examined the relationship between impaired mobility and the progressive accrual of WMSA. Mobility was assessed with the Short Physical Performance Battery (SPPB) and quantitative measures of gait and balance. Fourteen subjects had baseline and follow-up MRI scans performed 20 months apart. WMSA was detected and quantified using automated computer algorithms. In the control subjects, WMSA volume increased by 0.02±0.05% ICCV (percent intracranial cavity volume)/year while the WMSA of mobility impaired subjects increased five-times faster (0.10±0.10 ICCV/year, p=0.03). WMSA volume was related to some of the mobility measures and was sensitive to change which was not true of the other MRI variables. The study demonstrates the sensitivity of longitudinal automated volumetric analysis of WMSA to differentiate differences in the accrual rate of WMSA in groups selected on the basis of mobility. Based on these results, we propose that a subset of subjects with mobility impairment have accelerated, disease related WMSA accrual, thus explaining the rapid progression of mobility impairment in some older persons without apparent cause. This study demonstrates that quantitative MRI and performance measures can provide valuable insight into the rate of progression and pathophysiologic abnormalities underlying mobility impairment.

Automated detection of diabetic retinopathy on digital fundus images

The aim was to develop an automated screening system to analyse digital colour retinal images for important features of non-proliferative diabetic retinopathy (NPDR). High performance pre-processing of the colour images was performed. Previously described automated image analysis systems were used to detect major landmarks of the retinal image (optic disc, blood vessels and fovea). Recursive region growing segmentation algorithms combined with the use of a new technique, termed a 'Moat Operator', were used to automatically detect features of NPDR. These features included haemorrhages and microaneurysms (HMA), which were treated as one group, and hard exudates as another group. Sensitivity and specificity data were calculated by comparison with an experienced fundoscopist. The algorithm for exudate recognition was applied to 30 retinal images of which 21 contained exudates and nine were without pathology. The sensitivity and specificity for exudate detection were 88.5% and 99.7%, respectively, when compared with the ophthalmologist. HMA were present in 14 retinal images. The algorithm achieved a sensitivity of 77.5% and specificity of 88.7% for detection of HMA. Fully automated computer algorithms were able to detect hard exudates and HMA. This paper presents encouraging results in automatic identification of important features of NPDR.

Enhancement of protein modeling by human intervention in applying the automatic programs 3D-JIGSAW and 3D-PSSM.

Fourteen models were constructed and analyzed for the comparative modeling section of Critical Assessment of Techniques for Protein Structure Prediction (CASP4). Sequence identity between each target and the best possible parent(s) ranged between 55 and 13%, and the root-mean-square deviation between model and target was from 0.8 to 17.9 A. In the fold recognition section, 10 of the 11 remote homologues were recognized. The modeling protocols are a combination of automated computer algorithms, 3D-JIGSAW (for comparative modeling) and 3D-PSSM (for fold recognition), with human intervention at certain critical stages. In particular, intervention is required to check superfamily assignment, best possible parents from which to model, sequence alignments to those parents and take-off regions for modeling variable regions. There now is a convergence of algorithms for comparative modeling and fold recognition, particularly in the region of remote homology.

Model building by comparison at CASP3: Using expert knowledge and computer automation

Ten models were constructed for the comparative modeling section of the Critical Assessment of Techniques for Protein Structure Prediction-3 (CASP3). Sequence identity between each target and the best possible parent(s) ranged between 12% and 64%. The modeling protocol is a mixture of automated computer algorithms with human intervention at certain critical stages. In particular, intervention is required to check sequence alignments and the selection of parameters for various computer programs. Seven of the targets were constructed from single-parent templates, and three were constructed from multiple parents. The reasons for such a high ratio of modeling from single parents only are discussed. Models constructed from multiple parents were found to be more accurate than models constructed from single parents only. A novel loop-modeling algorithm is presented that consists of fragment database searches, several fragment libraries, and mean-field calculations on representative fragment candidates.

Model building by comparison at CASP3: Using expert knowledge and computer automation

Ten models were constructed for the comparative modeling section of the Critical Assessment of Techniques for Protein Structure Prediction-3 (CASP3). Sequence identity between each target and the best possible parent(s) ranged between 12% and 64%. The modeling protocol is a mixture of automated computer algorithms with human intervention at certain critical stages. In particular, intervention is required to check sequence alignments and the selection of parameters for various computer programs. Seven of the targets were constructed from single-parent templates, and three were constructed from multiple parents. The reasons for such a high ratio of modeling from single parents only are discussed. Models constructed from multiple parents were found to be more accurate than models constructed from single parents only. A novel loop-modeling algorithm is presented that consists of fragment database searches, several fragment libraries, and mean-field calculations on representative fragment candidates. Proteins Suppl 1999;3:47–54. © 1999 Wiley-Liss, Inc.

A completely automated activation-repolarization interval algorithm for directly coupled unipolar electrograms and its three-dimensional correlation with refractory periods

Sintered AgAgCl needle electrodes were used to record unipolar directly coupled (DC) electrograms transmurally from right and left ventricular myocardium. Direct coupling yields unfiltered low-frequency signal content, preserving repolarization waveform morphology. These electrograms were analyzed using completely automated computer algorithms for the timing of local activation and local repolarization that permit beat-to-beat determinations of these parameters. The local activation to repolarization interval was compared with the refractory period obtained with the extrastimulus technique from the same electrode during varied basic cycle lengths and pacing sequences. Pooling of the activation to refractory period versus activation-to-repolarization data from all sites and pacing sequences produced a correlation coefficient of 0.96, with a standard error of the estimate of 5.1 msec. The results demonstrate that DC-coupled unipolar electrograms may prove useful in direct, three-dimensional local repolarization timing and in simultaneous assessment of the temporal and spatial dispersion of changes in repolarization from multiple sites.