Vergence-Accommodation Conflict: Increased Presbyopia in Virtual Reality.

BACKGROUND: Performance and symptoms in completing a visual search task on a PC monitor and using a head-mounted display (HMD) were compared for different viewing conditions and between users of different ages. PATIENTS AND METHODS: Twenty-three young (M = 30 y, SD = 7 y) and 23 older (M = 52 y, SD = 5 y) participants performed a visual search task presented on a PC monitor. The task was repeated using an HMD for a near and a far virtual viewing distance. Reaction times (RT), detection sensitivity (d'), and symptoms were recorded for the three different viewing conditions. RESULTS: RT and d' were not affected by the viewing condition (p > 0.05). In contrast, symptoms significantly depended on the viewing condition but were, in part, not significantly affected by age. It is interesting to note that although not significant, young participants reported more ocular symptoms than older participants in the near vision task carried out using the HMD. DISCUSSION: HMD increases visual symptoms. However, HMD could be, in part, a remedy to problems when using visual aids for near work, in particular for presbyopes.

Positioning System of Infrared Sensors Based on ZnO Thin Film.

Infrared sensors incorporating suspended zinc oxide (ZnO) pyroelectric films and thermally insulated silicon substrates are fabricated using conventional MEMS-based thin-film deposition, photolithography, and etching techniques. The responsivity of the pyroelectric film is improved via annealing at 500 °C for 4 h. The voltage response of the fabricated sensors is evaluated experimentally for a substrate thickness of 1 µm over a sensing range of 30 cm. The results show that the voltage signal varies as an inverse exponential function of the distance. A positioning system based on three infrared sensors is implemented in LabVIEW. It is shown that the position estimates obtained using the proposed system are in excellent agreement with the actual locations. In general, the results presented in this study provide a useful source of reference for the further development of MEMS-based pyroelectric infrared sensors.

The effects of regular dental scaling on the complications and mortality after stroke: a retrospective cohort study based on a real-world database.

BACKGROUND: Previous observational studies have shown that people with dental scaling (DS) had decreased risk of stroke. However, limited information is available on the association between DS and poststroke outcomes. The present study aimed to evaluate the effects of regular DS on the complications and mortality after stroke. METHODS: We conducted a retrospective cohort study of 49,547 hospitalized stroke patients who received regular DS using 2010-2017 claims data of Taiwan's National Health Insurance. Using a propensity-score matching procedure, we selected 49,547 women without DS for comparison. Multiple logistic regressions were used to calculate the odds ratios (ORs) and 95% confidence intervals (CIs) of poststroke complications and in-hospital mortality associated with regular DS. RESULTS: Stroke patients with regular DS had significantly lower risks of poststroke pneumonia (OR 0.58, 95% CI 0.54-0.63), septicemia (OR 0.58, 95% CI 0.54-0.63), urinary tract infection (OR 0.68, 95% CI 0.66-0.71), intensive care (OR 0.81, 95% CI 0.78-0.84), and in-hospital mortality (OR 0.66, 95% CI 0.62-0.71) compared with non-DS stroke patients. Stroke patients with regular DS also had shorter hospital stays (p < 0.0001) and less medical expenditures (p < 0.0001) during stroke admission than the control group. Lower rates of poststroke adverse events in patients with regular DS were noted in both sexes, all age groups, and people with various types of stroke. CONCLUSION: Stroke patients with regular DS showed fewer complications and lower mortality compared with patients had no DS. These findings suggest the urgent need to promote regular DS for this susceptible population of stroke patients.

Graph Spatio-Temporal Networks for Manufacturing Sales Forecast and Prevention Policies in Pandemic Era.

Worldwide manufacturing industries are significantly affected by COVID-19 pandemic because of their production characteristics with low-cost country sourcing, globalization, and inventory level. To analyze the correlated time series, spatial-temporal model becomes more attractive, and the graph convolution network (GCN) is also commonly used to provide more information to the nodes and its neighbors in the graph. Recently, attention-adjusted graph spatio-temporal network (AGSTN) was proposed to address the problem of pre-defined graph in GCN by combining multi-graph convolution and attention adjustment to learn spatial and temporal correlations over time. However, AGSTN may show potential problem with limited small non-sensor data; particularly, convergence issue. This study proposes several variants of AGSTN and applies them to non-sensor data. We suggest data augmentation and regularization techniques such as edge selection, time series decomposition, prevention policies to improve AGSTN. An empirical study of worldwide manufacturing industries in pandemic era was conducted to validate the proposed variants. The results show that the proposed variants significantly improve the prediction performance at least around 20% on mean squared error (MSE) and convergence problem.

Effect of synthesis time on plasmonic properties of Ag dendritic nanoforests.

The effects of synthesis time on the plasmonic properties of Ag dendritic nanoforests on Si substrate (Ag-DNF/Si) samples synthesized through the fluoride-assisted galvanic replacement reaction were investigated. The Ag-DNF/Si samples were characterized using scanning electron microscopy, energy-dispersive X-ray spectroscopy, reflection spectroscopy, X-ray diffraction and surface-enhanced Raman spectroscopy (SERS). The prolonged reaction time led to the growth of an Ag-DNF layer and etched Si hole array. SEM images and variations in the fractal dimension index indicated that complex-structure, feather-like leaves became coral-like branches between 30 and 60 min of synthesis. The morphological variation during the growth of the Ag DNFs resulted in different optical responses to light illumination, especially those of light harvest and energy transformation. The sample achieved the most desirable light-to-heat conversion efficiency and SERS response with a 30 min growth time. A longer synthesis time or thicker Ag-DNF layer on the Si substrate did not have superior plasmonic properties.

Image registration method using representative feature detection and iterative coherent spatial mapping for infrared medical images with flat regions.

In the registration of medical images, nonrigid registration targets, images with large displacement caused by different postures of the human body, and frequent variations in image intensity due to physiological phenomena are substantial problems that make medical images less suitable for intensity-based image registration modes. These problems also greatly increase the difficulty and complexity of feature detection and matching for feature-based image registration modes. This research introduces an automatic image registration algorithm for infrared medical images that offers the following benefits: effective detection of feature points in flat regions (cold patterns) that appear due to changes in the human body's thermal patterns, improved mismatch removal through coherent spatial mapping for improved feature point matching, and large-displacement optical flow for optimal transformation. This method was compared with various classical gold standard image registration methods to evaluate its performance. The models were compared for the three key steps of the registration process-feature detection, feature point matching, and image transformation-and the results are presented visually and quantitatively. The results demonstrate that the proposed method outperforms existing methods in all tasks, including in terms of the features detected, uniformity of feature points, matching accuracy, and control point sparsity, and achieves optimal image transformation. The performance of the proposed method with four common image types was also evaluated, and the results verify that the proposed method has a high degree of stability and can effectively register medical images under a variety of conditions.

ADAM Challenge: Detecting Age-Related Macular Degeneration From Fundus Images.

Age-related macular degeneration (AMD) is the leading cause of visual impairment among elderly in the world. Early detection of AMD is of great importance, as the vision loss caused by this disease is irreversible and permanent. Color fundus photography is the most cost-effective imaging modality to screen for retinal disorders. Cutting edge deep learning based algorithms have been recently developed for automatically detecting AMD from fundus images. However, there are still lack of a comprehensive annotated dataset and standard evaluation benchmarks. To deal with this issue, we set up the Automatic Detection challenge on Age-related Macular degeneration (ADAM), which was held as a satellite event of the ISBI 2020 conference. The ADAM challenge consisted of four tasks which cover the main aspects of detecting and characterizing AMD from fundus images, including detection of AMD, detection and segmentation of optic disc, localization of fovea, and detection and segmentation of lesions. As part of the ADAM challenge, we have released a comprehensive dataset of 1200 fundus images with AMD diagnostic labels, pixel-wise segmentation masks for both optic disc and AMD-related lesions (drusen, exudates, hemorrhages and scars, among others), as well as the coordinates corresponding to the location of the macular fovea. A uniform evaluation framework has been built to make a fair comparison of different models using this dataset. During the ADAM challenge, 610 results were submitted for online evaluation, with 11 teams finally participating in the onsite challenge. This paper introduces the challenge, the dataset and the evaluation methods, as well as summarizes the participating methods and analyzes their results for each task. In particular, we observed that the ensembling strategy and the incorporation of clinical domain knowledge were the key to improve the performance of the deep learning models.

Automated segmentation of normal and diseased coronary arteries - The ASOCA challenge.

Cardiovascular disease is a major cause of death worldwide. Computed Tomography Coronary Angiography (CTCA) is a non-invasive method used to evaluate coronary artery disease, as well as evaluating and reconstructing heart and coronary vessel structures. Reconstructed models have a wide array of for educational, training and research applications such as the study of diseased and non-diseased coronary anatomy, machine learning based disease risk prediction and in-silico and in-vitro testing of medical devices. However, coronary arteries are difficult to image due to their small size, location, and movement, causing poor resolution and artefacts. Segmentation of coronary arteries has traditionally focused on semi-automatic methods where a human expert guides the algorithm and corrects errors, which severely limits large-scale applications and integration within clinical systems. International challenges aiming to overcome this barrier have focussed on specific tasks such as centreline extraction, stenosis quantification, and segmentation of specific artery segments only. Here we present the results of the first challenge to develop fully automatic segmentation methods of full coronary artery trees and establish the first large standardized dataset of normal and diseased arteries. This forms a new automated segmentation benchmark allowing the automated processing of CTCAs directly relevant for large-scale and personalized clinical applications.

Added Value of Computed Tomography Virtual Intravascular Endoscopy in the Evaluation of Coronary Arteries with Stents or Plaques.

Coronary computed tomography angiography (CCTA) is a widely used imaging modality for diagnosing coronary artery disease (CAD) but is limited by a high false positive rate when evaluating coronary arteries with stents and heavy calcifications. Virtual intravascular endoscopy (VIE) images generated from CCTA can be used to qualitatively assess the vascular lumen and might be helpful for overcoming this challenge. In this study, one hundred subjects with coronary stents underwent both CCTA and invasive coronary angiography (ICA). A total of 902 vessel segments were analyzed using CCTA and VIE. The vessel segments were first analyzed on CCTA alone. Then, using VIE, the segments were classified qualitatively as either negative or positive for in-stent restenosis (ISR) or CAD. These results were compared, using ICA as the reference, to determine the added diagnostic value of VIE. Of the 902 analyzed vessel segments, CCTA/VIE had sensitivity, specificity, accuracy, positive predictive value, and negative predictive value (shown in %) of 93.9/90.2, 96.2/98.2, 96.0/97.7, 70.0/83.1, and 99.4/99.0, respectively, in diagnosing ISR or CAD, with significantly improved specificity (p = 0.025), accuracy (p = 0.046), and positive predictive value (p = 0.047). VIE can be a helpful addition to CCTA when evaluating coronary arteries.

Design and Application of MEMS-Based Hall Sensor Array for Magnetic Field Mapping.

A magnetic field measurement system based on an array of Hall sensors is proposed. The sensors are fabricated using conventional microelectromechanical systems (MEMS) techniques and consist of a P-type silicon substrate, a silicon dioxide isolation layer, a phosphide-doped cross-shaped detection zone, and gold signal leads. When placed within a magnetic field, the interaction between the local magnetic field produced by the working current and the external magnetic field generates a measurable Hall voltage from which the strength of the external magnetic field is then derived. Four Hall sensors are fabricated incorporating cross-shaped detection zones with an identical aspect ratio (2.625) but different sizes (S, M, L, and XL). For a given working current, the sensitivities and response times of the four devices are found to be almost the same. However, the offset voltage increases with the increasing size of the detection zone. A 3 × 3 array of sensors is assembled into a 3D-printed frame and used to determine the magnetic field distributions of a single magnet and a group of three magnets, respectively. The results show that the constructed 2D magnetic field contour maps accurately reproduce both the locations of the individual magnets and the distributions of the magnetic fields around them.

SNP Data Science for Classification of Bipolar Disorder I and Bipolar Disorder II.

Bipolar disorder I (BD-I) and bipolar disorder II (BD-II) have specific characteristics and clear diagnostic criteria, but quite different treatment guidelines. In clinical practice, BD-II is commonly mistaken as a mild form of BD-I. This study uses data science technique to identify the important Single Nucleotide Polymorphisms (SNPs) significantly affecting the classifications of BD-I and BD-II, and develops a set of complementary diagnostic classifiers to enhance the diagnostic process. Screening assessments and SNP genotypes of 316 Han Chinese were performed with the Affymetrix Axiom Genome-Wide TWB Array Plate. The results show that the classifier constructed by 23 SNPs reached the area under curve of ROC (AUC) level of 0.939, while the classifier constructed by 42 SNPs reached the AUC level of 0.9574, which is a mere addition of 1.84 percent. The accuracy rate of classification increased by 3.46 percent. This study also uses Gene Ontology (GO) and Pathway to conduct a functional analysis and identify significant items, including calcium ion binding, GABA-A receptor activity, Rap1 signaling pathway, ECM proteoglycans, IL12-mediated signaling events, Nicotine addiction), and PI3K-Akt signaling pathway. The study can address time-consuming SNPs identification and also quantify the effect of SNP-SNP interactions.

Risk of dementia in patients with periodontitis and related protective factors: A nationwide retrospective cohort study.

AIM: We aimed to evaluate the long-term risk of dementia in patients with periodontitis and its associated factors. MATERIALS AND METHODS: Using Taiwan's National Health Insurance Database, we identified 56,018 patients aged ≥50 years with newly diagnosed periodontitis in 2000-2008. A cohort of 56,018 adults without periodontitis was selected for comparison, with matching by age and sex. Both cohorts were followed from 2000 to the end of 2013, and incident dementia was identified during the follow-up period. Adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) of dementia associated with periodontitis were calculated in the multivariate regressions. RESULTS: Periodontitis was associated with dementia risk (HR: 1.79, 95% CI: 1.67-1.93), and the association between periodontitis and dementia risk was significant in men, women, and people aged more than 60 years. Among patients with periodontitis, the use of statins (HR: 0.78, 95% CI: 0.71-0.87), metformin (HR: 0.53, 95% CI: 0.44-0.62), and influenza vaccination (HR: 0.67, 95% CI: 0.61-0.74) were associated with a reduced risk of dementia, while diabetes, mental disorders, and stroke were major significant risk factors. CONCLUSIONS: Periodontitis was a risk factor for dementia, while the use of statins and metformin may reduce the risk of dementia.

Fully automated lesion segmentation and visualization in automated whole breast ultrasound (ABUS) images.

BACKGROUND: The number of breast cancer patients has increased each year, and the demand for breast cancer detection has become quite large. There are many common breast cancer diagnostic tools. The latest automated whole breast ultrasound (ABUS) technology can obtain a complete breast tissue structure, which improves breast cancer detection technology. However, due to the large amount of ABUS image data, manual interpretation is time-consuming and labor-intensive. If there are lesions in multiple images, there may be some omissions. In addition, if further volume information or the three-dimensional shape of the lesion is needed for therapy, it is necessary to manually segment each lesion, which is inefficient for diagnosis. Therefore, automatic lesion segmentation for ABUS is an important issue for guiding therapy. METHODS: Due to the amount of speckle noise in an ultrasonic image and the low contrast of the lesion boundary, it is quite difficult to automatically segment the lesion. To address the above challenges, this study proposes an automated lesion segmentation algorithm. The architecture of the proposed algorithm can be divided into four parts: (I) volume of interest selection, (II) preprocessing, (III) segmentation, and (IV) visualization. A volume of interest (VOI) is automatically selected first via a three-dimensional level-set, and then the method uses anisotropic diffusion to address the speckled noise and intensity inhomogeneity correction to eliminate shadowing artifacts before the adaptive distance regularization level set method (DRLSE) conducts segmentation. Finally, the two-dimensional segmented images are reconstructed for visualization in the three-dimensional space. RESULTS: The ground truth is delineated by two radiologists with more than 10 years of experience in breast sonography. In this study, three performance assessments are carried out to evaluate the effectiveness of the proposed algorithm. The first assessment is the similarity measurement. The second assessment is the comparison of the results of the proposed algorithm and the Chan-Vese level set method. The third assessment is the volume estimation of phantom cases. In this study, in the 2D validation of the first assessment, the area Dice similarity coefficients of the real cases named cases A, real cases B and phantoms are 0.84±0.02, 0.86±0.03 and 0.92±0.02, respectively. The overlap fraction (OF) and overlap value (OV) of the real cases A are 0.84±0.06 and 0.78±0.04, real case B are 0.91±0.04 and 0.82±0.05, respectively. The overlap fraction (OF) and overlap value (OV) of the phantoms are 0.95±0.02 and 0.92±0.03, respectively. In the 3D validation, the volume Dice similarity coefficients of the real cases A, real cases B and phantoms are 0.85±0.02, 0.89±0.04 and 0.94±0.02, respectively. The overlap fraction (OF) and overlap value (OV) of the real cases A are 0.82±0.06 and 0.79±0.04, real cases B are 0.92±0.04 and 0.85±0.07, respectively. The overlap fraction (OF) and overlap value (OV) of the phantoms are 0.95±0.01 and 0.93±0.04, respectively. Therefore, the proposed algorithm is highly reliable in most cases. In the second assessment, compared with Chan-Vese level set method, the Dice of the proposed algorithm in real cases A, real cases B and phantoms are 0.84±0.02, 0.86±0.03 and 0.92±0.02, respectively. The Dice of Chan-Vese level set in real cases A, real cases B and phantoms are 0.65±0.23, 0.69±0.14 and 0.76±0.14, respectively. The Dice performance of different methods on segmentation shows a highly significant impact (P<0.01). The results show that the proposed algorithm is more accurate than Chan-Vese level set method. In the third assessment, the Spearman's correlation coefficient between the segmented volumes and the corresponding ground truth volumes is ρ=0.929 (P=0.01). CONCLUSIONS: In summary, the proposed method can batch process ABUS images, segment lesions, calculate their volumes and visualize lesions to facilitate observation by radiologists and physicians.

Data Science for Extubation Prediction and Value of Information in Surgical Intensive Care Unit.

Besides the traditional indices such as biochemistry, arterial blood gas, rapid shallow breathing index (RSBI), acute physiology and chronic health evaluation (APACHE) II score, this study suggests a data science framework for extubation prediction in the surgical intensive care unit (SICU) and investigates the value of the information our prediction model provides. A data science framework including variable selection (e.g., multivariate adaptive regression splines, stepwise logistic regression and random forest), prediction models (e.g., support vector machine, boosting logistic regression and backpropagation neural network (BPN)) and decision analysis (e.g., Bayesian method) is proposed to identify the important variables and support the extubation decision. An empirical study of a leading hospital in Taiwan in 2015-2016 is conducted to validate the proposed framework. The results show that APACHE II and white blood cells (WBC) are the two most critical variables, and then the priority sequence is eye opening, heart rate, glucose, sodium and hematocrit. BPN with selected variables shows better prediction performance (sensitivity: 0.830; specificity: 0.890; accuracy 0.860) than that with APACHE II or RSBI. The value of information is further investigated and shows that the expected value of experimentation (EVE), 0.652 days (patient staying in the ICU), is saved when comparing with current clinical experience. Furthermore, the maximal value of information occurs in a failure rate around 7.1% and it reveals the "best applicable condition" of the proposed prediction model. The results validate the decision quality and useful information provided by our predicted model.

Automated drosophila heartbeat counting based on image segmentation technique on optical coherence tomography.

Drosophila and human cardiac genes are very similar. Biological parametric studies on drosophila cardiac have improved our understanding of human cardiovascular disease. Drosophila cardiac consist of five circular chambers: a conical chamber (CC) and four ostia sections (O1-O4). Due to noise and grayscale discontinuity on optical coherence tomography (OCT) images, previous researches used manual counting or M-mode to analyze heartbeats, which are inefficient and time-consuming. An automated drosophila heartbeat counting algorithm based on the chamber segmentation is developed for OCT in this study. This algorithm has two parts: automated chamber segmentation and heartbeat counting. In addition, this study proposes a principal components analysis (PCA)-based supervised learning method for training the chamber contours to make chamber segmentation more accurate. The mean distances between the conical, second and third chambers attained by the proposed algorithm and the corresponding manually delineated boundaries defined by two experts were 1.26 ± 0.25, 1.47 ± 1.25 and 0.84 ± 0.60 (pixels), respectively. The area overlap similarities were 0.83 ± 0.09, 0.75 ± 0.11 and 0.74 ± 0.12 (pixels), respectively. The average calculated heart rates of two-week and six-week drosophila were about 4.77 beats/s and 4.73 beats/s, respectively, which was consistent with the results of manual counting.

Decentralized allocation of emission permits by Nash data envelopment analysis in the coal-fired power market.

Allocation of emission permits (AEP) provides valuable guidelines to support environmental regulatory policies for pollutant emission, in particular, CO2 as the key contributors to climate change. Most of previous studies in literature developed the centralized AEP model and focused on the coal-fired power market, one of the main sources of air pollution. However, the power market is usually imperfectly competitive and some of them are gradually deregulated, this justifies the motivation of developing a decentralized AEP model. This study proposes a decentralized AEP model which suggests Nash equilibrium as an allocatively efficient benchmark in an imperfectly competitive market with endogenous price. The proposed model is formulated by data envelopment analysis (DEA) and transformed into the mixed complementarity problem (MiCP) for identifying the Nash equilibrium. A study of coal-fired power plants operating in China in 2013 is conducted and the results show that the decentralized model complements the centralized model; in particular, with considering the price and market structure, the proposed decentralized model described in this study investigates the potential for efficiency improvement after AEP among the coal-fired plants.

Cross-section and feasibility study on the non-invasive evaluation of muscle hemodynamic responses in Duchenne muscular dystrophy by using a near-infrared diffuse optical technique.

Duchenne muscular dystrophy (DMD) is an X-linked debilitating muscular disease that may decrease nitric oxide (NO) production and lead to functional muscular ischemia. Currently, the 6-minute walk test (6-MWT) and the North Star Ambulatory Assessment (NSAA) are the primary outcome measures in clinical trials, but they are severely limited by the subjective consciousness and mood of patients, and can only be used in older and ambulatory boys. This study proposed using functional near-infrared spectroscopy (fNIRS) to evaluate the dynamic changes in muscle hemodynamic responses (gastrocnemius and forearm muscle) during a 6-MWT and a venous occlusion test (VOT), respectively. Muscle oxygenation of the forearm was evaluated non-invasively before, during and after VOT in all participants (included 30 DMD patients and 30 age-matched healthy controls), while dynamic muscle oxygenation of gastrocnemius muscle during 6-MWT was determined in ambulatory participants (n = 18) and healthy controls (n = 30). The results reveal that impaired muscle oxygenation was observed during 6-MWT in DMD patients that may explain why the DMD patients walked shorter distances than healthy controls. Moreover, the results of VOT implied that worsening muscle function was associated with a lower supply of muscle oxygenation and may provide useful information on the relationship between muscular oxygen consumption and supply for the clinical diagnosis of DMD. Therefore, the method of fNIRS with VOT possesses great potential in future evaluations of DMD patients that implies a good feasibility for clinical application such as for monitoring disease severity of DMD.

Well-aligned Vertically Oriented ZnO Nanorod Arrays and their Application in Inverted Small Molecule Solar Cells.

This manuscript describes how to design and fabricate efficient inverted solar cells, which are based on a two-dimensional conjugated small molecule (SMPV1) and [6,6]-phenyl-C71-butyric acid methyl ester (PC71BM), by utilizing ZnO nanorods (NRs) grown on a high quality Al-doped ZnO (AZO) seed layer. The inverted SMPV1:PC71BM solar cells with ZnO NRs that grew on both a sputtered and sol-gel processed AZO seed layer are fabricated. Compared with the AZO thin film prepared by the sol-gel method, the sputtered AZO thin film exhibits better crystallization and lower surface roughness, according to X-ray diffraction (XRD) and atomic force microscope (AFM) measurements. The orientation of the ZnO NRs grown on a sputtered AZO seed layer shows better vertical alignment, which is beneficial for the deposition of the subsequent active layer, forming better surface morphologies. Generally, the surface morphology of the active layer mainly dominates the fill factor (FF) of the devices. Consequently, the well-aligned ZnO NRs can be used to improve the carrier collection of the active layer and to increase the FF of the solar cells. Moreover, as an anti-reflection structure, it can also be utilized to enhance the light harvesting of the absorption layer, with the power conversion efficiency (PCE) of solar cells reaching 6.01%, higher than the sol-gel based solar cells with an efficiency of 4.74%.

An automated skin segmentation of Breasts in Dynamic Contrast-Enhanced Magnetic Resonance Imaging.

Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is used to diagnose breast disease. Obtaining anatomical information from DCE-MRI requires the skin be manually removed so that blood vessels and tumors can be clearly observed by physicians and radiologists; this requires considerable manpower and time. We develop an automated skin segmentation algorithm where the surface skin is removed rapidly and correctly. The rough skin area is segmented by the active contour model, and analyzed in segments according to the continuity of the skin thickness for accuracy. Blood vessels and mammary glands are retained, which remedies the defect of removing some blood vessels in active contours. After three-dimensional imaging, the DCE-MRIs without the skin can be used to see internal anatomical information for clinical applications. The research showed the Dice's coefficients of the 3D reconstructed images using the proposed algorithm and the active contour model for removing skins are 93.2% and 61.4%, respectively. The time performance of segmenting skins automatically is about 165 times faster than manually. The texture information of the tumors position with/without the skin is compared by the paired t-test yielded all p < 0.05, which suggested the proposed algorithm may enhance observability of tumors at the significance level of 0.05.

Is Intensity Inhomogeneity Correction Useful for Classification of Breast Cancer in Sonograms Using Deep Neural Network?

The sonogram is currently an effective cancer screening and diagnosis way due to the convenience and harmlessness in humans. Traditionally, lesion boundary segmentation is first adopted and then classification is conducted, to reach the judgment of benign or malignant tumor. In addition, sonograms often contain much speckle noise and intensity inhomogeneity. This study proposes a novel benign or malignant tumor classification system, which comprises intensity inhomogeneity correction and stacked denoising autoencoder (SDAE), and it is suitable for small-size dataset. A classifier is established by extracting features in the multilayer training of SDAE; automatic analysis of imaging features by the deep learning algorithm is applied on image classification, thus allowing the system to have high efficiency and robust distinguishing. In this study, two kinds of dataset (private data and public data) are used for deep learning models training. For each dataset, two groups of test images are compared: the original images and the images after intensity inhomogeneity correction, respectively. The results show that when deep learning algorithm is applied on the sonograms after intensity inhomogeneity correction, there is a significant increase of the tumor distinguishing accuracy. This study demonstrated that it is important to use preprocessing to highlight the image features and further give these features for deep learning models. In this way, the classification accuracy will be better to just use the original images for deep learning.