Automated gastric slow wave cycle partitioning and visualization for high-resolution activation time maps

High-resolution (HR) multi-electrode mapping has become an important technique for evaluating gastrointestinal (GI) slow wave (SW) behaviors. However, the application and uptake of HR mapping has been constrained by the complex and laborious task of analyzing the large volumes of retrieved data. Recently, a rapid and reliable method for automatically identifying activation times (ATs) of SWs was presented, offering substantial efficiency gains. To extend the automated data-processing pipeline, novel automated methods are needed for partitioning identified ATs into their propagation cycles, and for visualizing the HR spatiotemporal maps. A novel cycle partitioning algorithm (termed REGROUPS) is presented. REGROUPS employs an iterative REgion GROwing procedure and incorporates a Polynomial-surface-estimate Stabilization step, after initiation by an automated seed selection process. Automated activation map visualization was achieved via an isochronal contour mapping algorithm, augmented by a heuristic 2-step scheme. All automated methods were collectively validated in a series of experimental test cases of normal and abnormal SW propagation, including instances of patchy data quality. The automated pipeline performance was highly comparable to manual analysis, and outperformed a previously proposed partitioning approach. These methods will substantially improve the efficiency of GI HR mapping research.     

Automated detection of malignant features in confocal microscopy on superficial spreading melanoma versus nevi

In-vivo reflectance confocal microscopy (RCM) shows promise for the early detection of superficial spreading melanoma (SSM). RCM of SSM shows pagetoid melanocytes (PMs) in the epidermis and disarray at the dermal-epidermal junction (DEJ), which are automatically quantified with a computer algorithm that locates depth of the most superficial pigmented surface [D(SPS)(x,y)] containing PMs in the epidermis and pigmented basal cells near the DEJ. The algorithm uses 200 noninvasive confocal optical sections that image the superficial 200 μm of ten skin sites: five unequivocal SSMs and five nevi. The pattern recognition algorithm automatically identifies PMs in all five SSMs and finds none in the nevi. A large mean gradient ψ (roughness) between laterally adjacent points on D(SPS)(x,y) identifies DEJ disruption in SSM ψ = 11.7 ± 3.7 [-] for n = 5 SSMs versus a small ψ = 5.5 ± 1.0 [-] for n = 5 nevi (significance, p = 0.0035). Quantitative endpoint metrics for malignant characteristics make digital RCM data an attractive diagnostic asset for pathologists, augmenting studies thus far, which have relied largely on visual assessment.     

Automated detection of ocular focus

We characterize objectively the state of focus of the human eye, utilizing a bull's eye photodetector to detect the double-pass blur produced from a point source of light. A point fixation source of light illuminates the eye. Fundus-reflected light is focused by the optical system of the eye onto a bull's eye photodetector [consisting of an annulus (A) and a center (C) of approximately equal active area]. To generate focus curves, C/A is measured with a range of trial lenses in the light path. Three human eyes and a model eye are studied. In the model eye, the focus curve showed a sharp peak with a full width at half maximum (FWHM) of +/-0.25 D. In human eyes, the ratio C/A was >4 at best focus in all cases, with a FWHM of +/-1 D. The optical apparatus detects ocular focus (as opposed to refractive error) in real time. A device that can assess focus rapidly and objectively will make it possible to perform low-cost, mass screening for focusing problems such as may exist in children at risk for amblyopia.     

Automated misspelling detection and correction in clinical free-text records

Accurate electronic health records are important for clinical care and research as well as ensuring patient safety. It is crucial for misspelled words to be corrected in order to ensure that medical records are interpreted correctly. This paper describes the development of a spelling correction system for medical text. Our spell checker is based on Shannon’s noisy channel model, and uses an extensive dictionary compiled from many sources. We also use named entity recognition, so that names are not wrongly corrected as misspellings. We apply our spell checker to three different types of free-text data: clinical notes, allergy entries, and medication orders; and evaluate its performance on both misspelling detection and correction. Our spell checker achieves detection performance of up to 94.4% and correction accuracy of up to 88.2%. We show that high-performance spelling correction is possible on a variety of clinical documents.     

Automated detection of exudates for diabetic retinopathy screening

Automated image analysis is being widely sought to reduce the workload required for grading images resulting from diabetic retinopathy screening programmes. The recognition of exudates in retinal images is an important goal for automated analysis since these are one of the indicators that the disease has progressed to a stage requiring referral to an ophthalmologist. Candidate exudates were detected using a multi-scale morphological process. Based on local properties, the likelihoods of a candidate being a member of classes exudate, drusen or background were determined. This leads to a likelihood of the image containing exudates which can be thresholded to create a binary decision. Compared to a clinical reference standard, images containing exudates were detected with sensitivity 95.0% and specificity 84.6% in a test set of 13,219 images of which 300 contained exudates. Depending on requirements, this method could form part of an automated system to detect images showing either any diabetic retinopathy or referable diabetic retinopathy.     

Automated syndrome detection in a set of clinical facial photographs

Computer systems play an important role in clinical genetics and are a routine part of finding clinical diagnoses but make it difficult to fully exploit information derived from facial appearance. So far, automated syndrome diagnosis based on digital, facial photographs has been demonstrated under study conditions but has not been applied in clinical practice. We have therefore investigated how well statistical classifiers trained on study data comprising 202 individuals affected by one of 14 syndromes could classify a set of 91 patients for whom pictures were taken under regular, less controlled conditions in clinical practice. We found a classification accuracy of 21% percent in the clinical sample representing a ratio of 3.0 over a random choice. This contrasts with a 60% accuracy or 8.5 ratio in the training data. Producing average images in both groups from sets of pictures for each syndrome demonstrates that the groups exhibit large phenotypic differences explaining discrepancies in accuracy. A broadening of the data set is suggested in order to improve accuracy in clinical practice. In order to further this goal, a software package is made available that allows application of the procedures and contributions toward an improved data set.     

Automated detection of transient ST-segment episodes in 24h electrocardiograms

A novel automated system is presented for improved detection of transient ischaemic and heart rate-related ST-segment episodes in ‘real-world’ 24 h ambulatory ECG data. Using a combination of traditional time-domain and Karhunen-Loève transform-based approaches, the detector derives QRS complex and ST-segment morphology feature vectors and, by mimicking human examination of feature-vector time series and their trends, tracks the time-varying ST-segment reference level owing to clinically unimportant, non-ischaemic causes, such as slow drifts, axis shifts and conduction changes. The detector estimates the slowly varying ST-segment level trend, identifies step changes in the time series and subtracts the ST-segment reference level thus obtained from the ST-segment level to obtain the ST-segment deviation time series, which are suitable for detection of ST-segment episodes. The detector was developed using the Long-term ST database containing 24h ambulatory ECG records with human-expert annotated transient ischaemic and heart rate-related ST-segment episodes. The average ST episode detection sensitivity/positive predictivity obtained when using the annotations of the annotation protocol B of the database were 78.9%/80.7%. Evaluation of the detector using the European Society of Cardiology ST-T database as a test database showed average ST episode detection sensitivity/positive predictivity of 81.3%/89.2%, which are better performances, comparable with those of the systems being developed using the European database.     

Automated detection of medication administration errors in neonatal intensive care

ObjectiveTo improve neonatal patient safety through automated detection of medication administration errors (MAEs) in high alert medications including narcotics, vasoactive medication, intravenous fluids, parenteral nutrition, and insulin using the electronic health record (EHR); to evaluate rates of MAEs in neonatal care; and to compare the performance of computerized algorithms to traditional incident reporting for error detection.MethodsWe developed novel computerized algorithms to identify MAEs within the EHR of all neonatal patients treated in a level four neonatal intensive care unit (NICU) in 2011 and 2012. We evaluated the rates and types of MAEs identified by the automated algorithms and compared their performance to incident reporting. Performance was evaluated by physician chart review.ResultsIn the combined 2011 and 2012 NICU data sets, the automated algorithms identified MAEs at the following rates: fentanyl, 0.4% (4 errors/1005 fentanyl administration records); morphine, 0.3% (11/4009); dobutamine, 0 (0/10); and milrinone, 0.3% (5/1925). We found higher MAE rates for other vasoactive medications including: dopamine, 11.6% (5/43); epinephrine, 10.0% (289/2890); and vasopressin, 12.8% (54/421). Fluid administration error rates were similar: intravenous fluids, 3.2% (273/8567); parenteral nutrition, 3.2% (649/20124); and lipid administration, 1.3% (203/15227). We also found 13 insulin administration errors with a resulting rate of 2.9% (13/456). MAE rates were higher for medications that were adjusted frequently and fluids administered concurrently. The algorithms identified many previously unidentified errors, demonstrating significantly better sensitivity (82% vs. 5%) and precision (70% vs. 50%) than incident reporting for error recognition.ConclusionsAutomated detection of medication administration errors through the EHR is feasible and performs better than currently used incident reporting systems. Automated algorithms may be useful for real-time error identification and mitigation.     

Automated detection of masses in mammograms by local adaptive thresholding

In this paper, an algorithm for detection of suspicious masses from mammographic images is presented. The proposed algorithm was tested on a database of 61 mammograms on which masses had previously been marked by experienced radiologists. Results show that the proposed method exhibits for mass detection, a sensitivity of 95.91%. The area under receiver operating characteristic (ROC) Az was 0.946 when enhancement of the original image was performed before detection and 0.938 otherwise. Furthermore in some cases, we could detect some masses that the radiologists were not able to mark out.     

利用边缘和表面特征实现脑图像的自动配准

本文中我们使用基于CT、MR和PET图像等值特征表面的配准算法对多模医学图像进行了配准研究，在CT、MR和PET的原始图像中提取等值特征表面，进行图像的几何对准，并对结果进行初步评估，同时对该算法的稳健性，搜索最近点策略和采样策略进行了研究，结果表明；这种方法能够达到亚像素级的配准精度，是一种稳健、高精度、全自动的配准方法。     

Automated detection of lung nodules in CT scans: preliminary results

We have developed a fully automated computerized method for the detection of lung nodules in helical computed tomography (CT) scans of the thorax. This method is based on two-dimensional and three-dimensional analyses of the image data acquired during diagnostic CT scans. Lung segmentation proceeds on a section-by-section basis to construct a segmented lung volume within which further analysis is performed. Multiple gray-level thresholds are applied to the segmented lung volume to create a series of thresholded lung volumes. An 18-point connectivity scheme is used to identify contiguous three-dimensional structures within each thresholded lung volume, and those structures that satisfy a volume criterion are selected as initial lung nodule candidates. Morphological and gray-level features are computed for each nodule candidate. After a rule-based approach is applied to greatly reduce the number of nodule candidates that corresponds to nonnodules, the features of remaining candidates are merged through linear discriminant analysis. The automated method was applied to a database of 43 diagnostic thoracic CT scans. Receiver operating characteristic (ROC) analysis was used to evaluate the ability of the classifier to differentiate nodule candidates that correspond to actual nodules from false-positive candidates. The area under the ROC curve for this categorization task attained a value of 0.90 during leave-one-out-by-case evaluation. The automated method yielded an overall nodule detection sensitivity of 70% with an average of 1.5 false-positive detections per section when applied to the complete 43-case database. A corresponding nodule detection sensitivity of 89% with an average of 1.3 false-positive detections per section was achieved with a subset of 20 cases that contained only one or two nodules per case.     

Automated detection of a prostate Ni-Ti stent in electronic portal images

Planning target volumes (PTV) in fractionated radiotherapy still have to be outlined with wide margins to the clinical target volume due to uncertainties arising from daily shift of the prostate position. A recently proposed new method of visualization of the prostate is based on insertion of a thermo-expandable Ni-Ti stent. The current study proposes a new detection algorithm for automated detection of the Ni-Ti stent in electronic portal images. The algorithm is based on the Ni-Ti stent having a cylindrical shape with a fixed diameter, which was used as the basis for an automated detection algorithm. The automated method uses enhancement of lines combined with a grayscale morphology operation that looks for enhanced pixels separated with a distance similar to the diameter of the stent. The images in this study are all from prostate cancer patients treated with radiotherapy in a previous study. Images of a stent inserted in a humanoid phantom demonstrated a localization accuracy of 0.4-0.7 mm which equals the pixel size in the image. The automated detection of the stent was compared to manual detection in 71 pairs of orthogonal images taken in nine patients. The algorithm was successful in 67 of 71 pairs of images. The method is fast, has a high success rate, good accuracy, and has a potential for unsupervised localization of the prostate before radiotherapy, which would enable automated repositioning before treatment and allow for the use of very tight PTV margins.     

Automated quantification and architectural pattern detection of hepatic fibrosis in NAFLD

Accurate detection and quantification of hepatic fibrosis remain essential for assessing the severity of non-alcoholic fatty liver disease (NAFLD) and its response to therapy in clinical practice and research studies. Our aim was to develop an integrated artificial intelligence-based automated tool to detect and quantify hepatic fibrosis and assess its architectural pattern in NAFLD liver biopsies. Digital images of the trichrome-stained slides of liver biopsies from patients with NAFLD and different severity of fibrosis were used. Two expert liver pathologists semi-quantitatively assessed the severity of fibrosis in these biopsies and using a web applet provided a total of 987 annotations of different fibrosis types for developing, training and testing supervised machine learning models to detect fibrosis. The collagen proportionate area (CPA) was measured and correlated with each of the pathologists semi-quantitative fibrosis scores. Models were created and tested to detect each of six potential fibrosis patterns. There was good to excellent correlation between CPA and the pathologist score of fibrosis stage. The coefficient of determination (R²) of automated CPA with the pathologist stages ranged from 0.60 to 0.86. There was considerable overlap in the calculated CPA across different fibrosis stages. For identification of fibrosis patterns, the models areas under the receiver operator curve were 78.6% for detection of periportal fibrosis, 83.3% for pericellular fibrosis, 86.4% for portal fibrosis and >90% for detection of normal fibrosis, bridging fibrosis, and presence of nodule/cirrhosis. In conclusion, an integrated automated tool could accurately quantify hepatic fibrosis and determine its architectural patterns in NAFLD liver biopsies.     

Spinal cord maps of spatiotemporal alpha-motoneuron activation in humans walking at different speeds

Functional MRI (fMRI) imaging of motoneuron activity in the human spinal cord is still in its infancy, and it will remain difficult to apply to walking. Here we present a viable alternative for documenting the spatiotemporal maps of alpha-motorneuron (MN) activity in the human spinal cord during walking, similar to the method recently reported for the cat. We recorded EMG activity from 16 to 32 ipsilateral limb and trunk muscles in 13 healthy subjects walking on a treadmill at different speeds (1-7 km/h) and mapped the recorded patterns onto the spinal cord in approximate rostrocaudal locations of the motoneuron pools. This approach can provide information about pattern generator output during locomotion in terms of segmental control rather than in terms of individual muscle control. A striking feature we found is that nearly every spinal segment undergoes at least two cycles of activation in the step cycle, thus supporting the idea of half-center oscillators controlling MN activation at any segmental level. The resulting spatiotemporal map patterns seem highly stereotyped over the range of walking speeds studied, although there were also some systematic redistributions of MN activity with speed. Bursts of MN activity were either temporally aligned across several spinal segments or switched between different segments. For example, the center of mass of MN activity in the lumbosacral levels generally shifted from rostral to caudal positions in two cycles for each step, revealing four major activation foci: two in the upper lumbar segments and two in the sacral segments. The results are consistent with the presence of at least two and possibly more pattern generators controlling the activation of lumbosacral MNs.     

Automated neural network detection of wavelet preprocessed electrocardiogram late potentials

The aim of the study is to investigate the potential of a feedforward neural network for detecting wavelet preprocessed late potentials. The terminal parts of a simulated QRS complex are processed with a continuous wavelet transform, which leads to a time-frequency represenation of the QRS complex. Then, diagnostic feature vectors are obtained by subdividing the representations into several regions and by processing the sum of the decomposition coefficients belonging to each region. The neural network is trained with these feature vectors. Simulated ECGs with varying signalto-noise ratios are used to train and test the classifier. Results show that correct classification ranges from 79% (high-level noise) to 99% (no noise). The study shows the potential of neural networks for the classification of late potentials that have been preprocessed by a wavelet transform. However, clinical use of this method still requires further investigation.     

Automated detection of gait initiation and termination using wearable sensors

This paper presents algorithms for detection of gait initiation and termination using wearable inertial measurement units and pressure-sensitive insoles. Body joint angles, joint angular velocities, ground reaction force and center of plantar pressure of each foot are obtained from these sensors and input into supervised machine learning algorithms. The proposed initiation detection method recognizes two events: gait onset (an anticipatory movement preceding foot lifting) and toe-off. The termination detection algorithm segments gait into steps, measures the signals over a buffer at the beginning of each step, and determines whether this measurement belongs to the final step. The approach is validated with 10 subjects at two gait speeds, using within-subject and subject-independent cross-validation. Results show that gait initiation can be detected timely and accurately, with few errors in the case of within-subject cross-validation and overall good performance in subject-independent cross-validation. Gait termination can be predicted in over 80% of trials well before the subject comes to a complete stop. Results also show that the two sensor types are equivalent in predicting gait initiation while inertial measurement units are generally superior in predicting gait termination. Potential use of the algorithms is foreseen primarily with assistive devices such as prostheses and exoskeletons.     

Basic features of phasic activation for reaching in vertical planes

The purpose of this study was to fully characterize the timing and intensity of the phasic portion of the electromyographic (EMG) waveform for reaching movements in vertical planes. Electromyographic activity was simultaneously recorded from nine superficial elbow and/or shoulder muscles while human subjects made rapid arm movements. Hand paths comprised 20 directions in a sagittal plane and 20 directions in a frontal plane. In order to focus on the more phasic aspects of muscle activation, estimates of postural EMG activity were subtracted from the EMG traces recorded during rapid reaches. These postural estimates were obtained from activity recorded during very slow reaches to the same targets. After subtraction of this postural activity, agonist or antagonist burst patterns were often observed in the phasic EMG traces. For nearly all muscles and all subjects, the relation between phasic EMG intensity and movement direction was a function with multiple peaks. For all muscles, the timing of phasic EMG bursts varied as a function of movement direction: the data from each muscle exhibited a gradual temporal shift of activity over a certain range of directions. This gradual temporal shift has no obvious correspondence to the mechanical requirements of the task and might represent a neuromuscular control strategy in which burst timing contributes to the specification of movement direction.     

Automated detection of toxigenic Clostridium difficile in clinical samples: isothermal tcdB amplification coupled to array-based detection

Clostridium difficile can carry a genetically variable pathogenicity locus (PaLoc), which encodes clostridial toxins A and B. In hospitals and in the community at large, this organism is increasingly identified as a pathogen. To develop a diagnostic test that combines the strengths of immunoassays (cost) and DNA amplification assays (sensitivity/specificity), we targeted a genetically stable PaLoc region, amplifying tcdB sequences and detecting them by hybridization capture. The assay employs a hot-start isothermal method coupled to a multiplexed chip-based readout, creating a manual assay that detects toxigenic C. difficile with high sensitivity and specificity within 1 h. Assay automation on an electromechanical instrument produced an analytical sensitivity of 10 CFU (95% probability of detection) of C. difficile in fecal samples, along with discrimination against other enteric bacteria. To verify automated assay function, 130 patient samples were tested: 31/32 positive samples (97% sensitive; 95% confidence interval [CI], 82 to 99%) and 98/98 negative samples (100% specific; 95% CI, 95 to 100%) were scored correctly. Large-scale clinical studies are now planned to determine clinical sensitivity and specificity.