Goal: In this work, we present an extensive description and evaluation of our method for blood vessel segmentation in fundus images based on a discriminatively trained fully connected conditional random field model. Methods: Standard segmentation priors such as a Potts model or total variation usually fail when dealing with thin and elongated structures. We overcome this difficulty by using a conditional random field model with more expressive potentials, taking advantage of recent results enabling inference of fully connected models almost in real time. Parameters of the method are learned automatically using a structured output support vector machine, a supervised technique widely used for structured prediction in a number of machine learning applications. Results: Our method, trained with state of the art features, is evaluated both quantitatively and qualitatively on four publicly available datasets: DRIVE, STARE, CHASEDB1, and HRF. Additionally, a quantitative comparison with respect to other strategies is included. Conclusion: The experimental results show that this approach outperforms other techniques when evaluated in terms of sensitivity, F1-score, G-mean, and Matthews correlation coefficient. Additionally, it was observed that the fully connected model is able to better distinguish the desired structures than the local neighborhood-based approach. Significance: Results suggest that this method is suitable for the task of segmenting elongated structures, a feature that can be exploited to contribute with other medical and biological applications.

/pdf/a-discriminatively-trained-fully-connected-conditional-29gv5ppedb.pdf

A Discriminatively Trained Fully Connected Conditional Random Field Model for Blood Vessel Segmentation in Fundus Images

Automatic segmentation of retinal blood vessels from fundus images plays an important role in the computer aided diagnosis of retinal diseases. The task of blood vessel segmentation is challenging due to the extreme variations in morphology of the vessels against noisy background. In this paper, we formulate the segmentation task as a multi-label inference task and utilize the implicit advantages of the combination of convolutional neural networks and structured prediction. Our proposed convolutional neural network based model achieves strong performance and significantly outperforms the state-of-the-art for automatic retinal blood vessel segmentation on DRIVE dataset with 95.33% accuracy and 0.974 AUC score.

A fully convolutional neural network based structured prediction approach towards the retinal vessel segmentation

Background: In India, poor feeding practices in early
childhood contribute to the burden of malnutrition and
infant and child mortality.
Objective. To estimate infant and young child feeding
indicators and determinants of selected feeding practices
in India.

Methods: The sample consisted of 20,108 children
aged 0 to 23 months from the National Family Health
Survey India 2005–06. Selected indicators were examined
against a set of variables using univariate and
multivariate analyses.

Results: Only 23.5% of mothers initiated breastfeeding
within the first hour after birth, 99.2% had ever
breastfed their infant, 89.8% were currently breastfeeding,
and 14.8% were currently bottle-feeding. Among
infants under 6 months of age, 46.4% were exclusively
breastfed, and 56.7% of those aged 6 to 9 months received
complementary foods. The risk factors for not exclusively
breastfeeding were higher household wealth index
quintiles (OR for richest = 2.03), delivery in a health
facility (OR = 1.35), and living in the Northern region.
Higher numbers of antenatal care visits were associated
with increased rates of exclusive breastfeeding (OR for
≥ 7 antenatal visits = 0.58). The rates of timely initiation
of breastfeeding were higher among women who were
better educated (OR for secondary education or above =
0.79), were working (OR = 0.79), made more antenatal
clinic visits (OR for ≥ 7 antenatal visits = 0.48), and
were exposed to the radio (OR = 0.76). The rates were
lower in women who were delivered by cesarean section
(OR = 2.52). The risk factors for bottle-feeding included
cesarean delivery (OR = 1.44), higher household wealth
index quintiles (OR = 3.06), working by the mother
(OR=1.29), higher maternal education level (OR=1.32),
urban residence (OR=1.46), and absence of postnatal
examination (OR=1.24). The rates of timely complementary
feeding were higher for mothers who had more
antenatal visits (OR=0.57), and for those who watched
television (OR=0.75).

Conclusions: Revitalization of the Baby Friendly Hospital
Initiative in health facilities is recommended. Targeted
interventions may be necessary to improve infant
feeding practices in mothers who reside in urban areas,
are more educated, and are from wealthier households.

Infant and young child feeding indicators and determinants of poor feeding practices in India : secondary data analysis of National Family Health Survey 2005–06

Retinal vessel segmentation is a key step towards the accurate visualization, diagnosis, early treatment and surgery planning of ocular diseases. For the last two decades, a tremendous amount of research has been dedicated in developing automated methods for segmentation of blood vessels from retinal fundus images. Despite the fact, segmentation of retinal vessels still remains a challenging task due to the presence of abnormalities, varying size and shape of the vessels, non-uniform illumination and anatomical variability between subjects. In this paper, we carry out a systematic review of the most recent advancements in retinal vessel segmentation methods published in last five years. The objectives of this study are as follows: first, we discuss the most crucial preprocessing steps that are involved in accurate segmentation of vessels. Second, we review most recent state-of-the-art retinal vessel segmentation techniques which are classified into different categories based on their main principle. Third, we quantitatively analyse these methods in terms of its sensitivity, specificity, accuracy, area under the curve and discuss newly introduced performance metrics in current literature. Fourth, we discuss the advantages and limitations of the existing segmentation techniques. Finally, we provide an insight into active problems and possible future directions towards building successful computer-aided diagnostic system.

Recent Advancements in Retinal Vessel Segmentation

Fuzzy based image edge detection algorithm for blood vessel detection in retinal images

Retinal fundus image plays an important role in the diagnosis of retinal related diseases. The detailed information of the retinal fundus image such as small vessels, microaneurysms, and exudates may be in low contrast, and retinal image enhancement usually gives help to analyze diseases related to retinal fundus image. Current image enhancement methods may lead to artificial boundaries, abrupt changes in color levels, and the loss of image detail. In order to avoid these side effects, a new retinal fundus image enhancement method is proposed. First, the original retinal fundus image was processed by the normalized convolution algorithm with a domain transform to obtain an image with the basic information of the background. Then, the image with the basic information of the background was fused with the original retinal fundus image to obtain an enhanced fundus image. Lastly, the fused image was denoised by a two-stage denoising method including the fourth order PDEs and the relaxed median filter. The retinal image databases, including the DRIVE database, the STARE database, and the DIARETDB1 database, were used to evaluate image enhancement effects. The results show that the method can enhance the retinal fundus image prominently. And, different from some other fundus image enhancement methods, the proposed method can directly enhance color images.

/pdf/retinal-fundus-image-enhancement-using-the-normalized-2lppwple9z.pdf

Retinal Fundus Image Enhancement Using the Normalized Convolution and Noise Removing

Background:Mothers are encouraged to exclusively breastfeed for the first 6 months. However, cesarean delivery rates have increased worldwide, which may affect breastfeeding.Research aim:This study...

Influences of Cesarean Delivery on Breastfeeding Practices and Duration: A Prospective Cohort Study.

Vessel segmentation in retinal fundus images is a preliminary step to clinical diagnosis for some systemic diseases and some eye diseases. The performances of existing methods for segmenting small vessels which are usually of more importance than the main vessels in a clinical diagnosis are not satisfactory in clinical use. In this paper, we present a method for both main and peripheral vessel segmentation. A local gray-level change enhancement algorithm called gray-voting is used to enhance the small vessels, while a two-dimensional Gabor wavelet is used to extract the main vessels. We fuse the gray-voting results with the 2D-Gabor filter results as pre-processing outcome. A Gaussian mixture model is then used to extract vessel clusters from the pre-processing outcome, while small vessels fragments are obtained using another gray-voting process, which complements the vessel cluster extraction already performed. At the last step, we eliminate the fragments that do not belong to the vessels based on the shape of the fragments. We evaluated the approach with two publicly available DRIVE (Staal et al., 2004) and STARE (Hoover et at., 2000) datasets with manually segmented results. For the STARE dataset, when using the second manually segmented results which include much more small vessels than the first manually segmented results as the “gold standard,” this approach achieved an average sensitivity, accuracy and specificity of 65.0%, 92.1% and 97.0%, respectively. The sensitivities of this approach were much higher than those of the other existing methods, with comparable specificities; these results thus demonstrated that this approach was sensitive to detection of small vessels.

A New Approach to Segment Both Main and Peripheral Retinal Vessels Based on Gray-Voting and Gaussian Mixture Model

Objective 
The aim of this study was to examine the mutual effect of pre-pregnancy body mass index (BMI), waist circumference (WC) and gestational weight gain (GWG) on obesity-related adverse pregnancy outcomes.


Methods 
This birth cohort study was conducted in three Streets in Changsha, China, including a total of 976 mother-child pairs. All data was collected within 15 days after deliveries from a self-administered questionnaire, maternal health manual and perinatal health care information system. Multivariate logistic regression models were conducted to estimate the effects of maternal pre-pregnancy BMI, WC and GWG on obesity-related adverse pregnancy outcomes including gestational diabetes mellitus (GDM), primary cesarean section (P-CS), large for gestational age (LGA) and composite outcome (one or more adverse pregnancy outcomes)


Results 
After controlling for all confounders, both maternal pre-pregnancy overweight/obesity and central adiposity contributed to increased risks of GDM [ORs 95% CIs = 2.19 (1.02–4.76) and 2.26 (1.11–4.60), respectively], P-CS [ORs 95% CIs = 1.66 (1.05–2.65) and 1.71 (1.11–2.63), respectively], LGA [ORs 95% CIs = 1.93 (1.07–3.50) and 2.14 (1.21–3.75), respectively] and composite outcome [ORs 95% CIs = 1.82 (1.15–2.87) and 1.98 (1.30–3.01), respectively] compared with mothers with normal pre-pregnancy weight and normal WC. Excessive GWG was found to be associated with an increased risk of LGA [OR 95% CI = 1.74 (1.05–2.89)], but was not significantly related to higher risks of GDM, P-CS and composite outcome [ORs 95% CIs = 0.90 (0.47–1.72), 1.08 (0.77–1.52), and 1.30 (0.94–1.79), respectively]. In terms of the joint effect of maternal pregestational BMI and WC on obesity-related composite outcome, mothers with both pre-pregnancy overweight and central adiposity had the highest risk of composite outcome [OR 95% CI = 3.96 (2.40–6.54)], compared with mothers without pre-pregnancy overweight or central adiposity.


Conclusions 
The results of this study suggest that maternal pre-pregnancy overweight/obesity and central adiposity may contribute to multiple obesity-related adverse pregnancy outcomes, excessive weight gain during pregnancy is associated with an increased risk of LGA. Healthcare providers should carry out health education, and guide women to keep an ideal BMI and WC prior to pregnancy and help them gain optimal weight during pregnancy based on their pre-pregnancy BMI and WC.

/pdf/the-mutual-effect-of-pre-pregnancy-body-mass-index-waist-4zftk55hog.pdf

The mutual effect of pre-pregnancy body mass index, waist circumference and gestational weight gain on obesity-related adverse pregnancy outcomes: A birth cohort study.

Breast milk is an ideal first food for infants in the initial months of life. This study aimed to identify modifiable individual factors in breastfeeding during the first six months of life in Hunan. This birth cohort study was conducted in three communities in Changsha, Hunan province. Data were collected through maternal health manuals and self-administered questionnaires during the follow-up period. To find risk factors and calculate odds ratios, generalized estimating equation models were applied. The final study included 951 mother-infant pairs. The prevalence of exclusive breastfeeding and any breastfeeding in this study was below the World Health Organization's breastfeeding goals. Infant age, passive smoking after delivery, minor postnatal depression, and feeding-bottles were shown to act negatively on both exclusive breastfeeding and any breastfeeding. In addition, cesarean delivery and delayed breastfeeding initiation had adverse effects on exclusive breastfeeding. Multiparas were less likely to continue any breastfeeding during the first few months. This study highlights the importance of screening probable maternal depression during primary care service and reducing secondhand smoke and feeding-bottle use. The results also suggest that healthcare professionals should provide more assistance and education to multiparas and mothers with cesarean delivery or delayed breastfeeding initiation.

https://www.mdpi.com/1660-4601/16/5/820/pdf

Ling Li

Papers

Retinal Fundus Image Enhancement Using the Normalized Convolution and Noise Removing

Influences of Cesarean Delivery on Breastfeeding Practices and Duration: A Prospective Cohort Study.

A New Approach to Segment Both Main and Peripheral Retinal Vessels Based on Gray-Voting and Gaussian Mixture Model

The mutual effect of pre-pregnancy body mass index, waist circumference and gestational weight gain on obesity-related adverse pregnancy outcomes: A birth cohort study.

Modifiable Individual Factors Associated with Breastfeeding: A Cohort Study in China.