Red lesions can be regarded as one of the earliest lesions in diabetic retinopathy (DR) and automatic detection of red lesions plays a critical role in diabetic retinopathy diagnosis. In this paper, a novel superpixel Multichannel Multifeature (MCMF) classification approach is proposed for red lesion detection. In this paper, firstly, a new candidate extraction method based on superpixel is proposed. Then, these candidates are characterized by multichannel features, as well as the contextual feature. Next, FDA classifier is introduced to classify the red lesions among the candidates. Finally, a postprocessing technique based on multiscale blood vessels detection is modified for removing nonlesions appearing as red. Experiments on publicly available DiaretDB1 database are conducted to verify the effectiveness of our proposed method.
Diabetic retinopathy (DR) is one of the most serious performances of diabetic and the majority of people suffering from diabetes mellitus for more than ten years will eventually develop DR [
There are several different components in retinal images (see Figure
A retinal image with different types of lesions and main anatomical features.
Numerous approaches have been proposed for red lesion detection. Among them, the earliest paper based on MA detection was proposed by Baudoin et al. [
A number of algorithms based on filtering have been proposed for red lesion detection in digital color fundus photographs. Quellec et al. [
Apart from the above-described detection approaches, several mathematical morphology based approaches have been proposed for the detection of red lesions. Júnior and Welfer [
All the above-mentioned detection approaches regard image pixels as the basic unit to distinguish red lesions from nonred lesions. However, image pixels are a consequence of the discrete representation of images but not natural entities. Compared to the pixel-based image representation, superpixel-based image representation is more consistent with human visual cognition and contains less redundancy [
To overcome the above-mentioned issues, in this paper, we propose a novel red lesion detection approach based on superpixel Multichannel Multifeature (MCMF). Two main contributions are as follows: on one hand, a novel candidate extraction scheme based on superpixel segmentation is given, which is more consistent with human visual cognition and contains less redundancy, improving efficiency and accuracy of subsequent image processing tasks. On the other hand, extensive features extracted from multiple-channel images have been proposed and introduced to our feature extraction process, which can improve the performance of red lesion detection.
Our proposed approach consists of the following five phases: first of all, preprocessing is used to make red lesions more visible. Next, candidates can be extracted by applying superpixel segmentation in digital color fundus photographs. And then, these candidates are characterized using not only intensity features in multichannel images (here, “multichannel” means that a series of images can be produced by different operations based on the original image), but also the contextual feature. Besides, Fisher Discriminant Analysis (FDA) [
The remainder of the paper is organized as follows. Section
In this section, we mainly introduce the proposed MCMF method for the detection of red lesions. Figure
A system description of the proposed approach.
The large luminosity, poor contrast, and noise always occur in retinal fundus images [
From left to right: original RGB color retinal images (a) and (c); enhancement images (b) and (d).
Given a preprocessed image
Region size means the size of superpixel segmentation, which plays a vital role in SLIC. For larger region sizes, spatial distances overweigh color proximity, causing the obtained superpixels to not adhere well to image boundaries. For smaller region sizes, the converse is true [
Retinal image segmentation using SLIC. (a) Original retinal fundus image; (b) detail from part of (a); (c)–(e) superpixel segmentation with region sizes 10, 30, and 50 pixels, respectively.
As for red lesion detection, some specific properties need to be considered. Firstly, some red lesions are very near the blood vessels, making it hard to distinguish them just in traditional RGB color space. Moreover, the appearance of red lesions is similar to the normal structures of the retinal, such as the blood vessels and the fovea, causing too much false positive (FP). Finally, the shape and size of red lesions especially for large hemorrhages are varied. Based on these facts, we adopt the strategy of extracting different features based on multichannel images for each candidate in this paper. The chosen multichannel images are listed as follows.
Multichannel images. (a) Original green channel image
For each of the above-described multichannel images, four kinds of statistic features including the maximum, minimum, mean, and median are extracted from each candidate. Besides, the total average intensity and standard deviation of the each preprocessing retinal image are also imported.
Since red lesions appear as darker regions with brighter surroundings, based on this characteristic, we also develop a novel and effective feature for distinguishing a candidate from its surroundings and background by mean intensity and the barycenter distance between the neighbor candidates. Here, let
Basically speaking, our proposed contextual feature needs to calculate the mean gray value of each candidate
In summary, thirty-one features are computed on each candidate and they can be represented as a vector in a 31-dimension feature set
In this section, Fisher Discriminant Analysis (FDA) [
Let
FDA transformation matrix
The above optimization problem can be regarded as the generalized eigenvalue problem below [
Since the projected class means
Given a new sample
After the classification stage, some nonlesions appearing as red such as blood vessels and fovea are often present and may be erroneously detected as red lesions. In order to avoid this problem, a postprocessing stage is incorporated into our approach for removing them and improving the robustness of the proposed method.
Since red lesions and blood vessel have the similar appearance, it is hard to distinguish them effectively. Besides, the red lesions cannot occur on the blood vessels [
Morphological opening
The high intensity structures can be eliminated by subtracting the CLAHE result image from
A series of morphological operations are used for blood vessels detection. (a)
Morphological opening with varying structuring elements is applied to
According to (
A binary vessel structure map
Repeat Steps
At last, we will obtain the final blood vessels map
Blood vessels extraction results. (a)–(d) show the different detection results with varying sizes of disc-shaped structuring elements 2, 3, 4, and 5, respectively; (e) the combination result of (a)–(d); (a1)–(d1) are the blood vessels map differences between (e) and (a)–(d), respectively.
The main difference between our proposed MSM blood vessels extraction method and the single scale blood vessels detection method [
The fovea appears as dark region located in the center of the macula region of the retina. Since the fovea region has a relative low intensity profile and its appearance is commonly much similar with the background causing it to be hard to distinguish from true red lesions. In order to improve the accuracy of the proposed method, the removal of fovea is indispensable. A method [
The final result of our proposed MCMF. (a) Original retinal image; (b) the corresponding ground truth; (c) final red lesions map; (d) overlaying red lesions map on original retinal image.
Figure
Results of our proposed red lesion detection approach on abnormal retinal images. (a) Original retinal images; (b) the corresponding red lesions maps.
In this section, we conduct extensive experiments to validate and evaluate the effectiveness of our proposed red lesion detection method on public DiaretDB1 retinal image database [
The DiaretDB1 database (Standard Diabetic Retinopathy Database Calibration level 1, version 1) [
We take two evaluation criterions to verify the effectiveness of our proposed red lesion detection method, such as sensitivity and specificity. These measures are calculated based on the given red lesion ground truth, which can be seen from the following:
In our experiments, we employ the Receiver Operating Characteristics (ROC) curve to evaluate the effectiveness of the proposed red lesion detection method. An ROC curve is the plot of sensitivity on the vertical axis and (
In this section, we will carry out three experiments based on DiaretDB1 dataset to verify the effectiveness of our proposed method. In our experiments, we employ two different kinds of criteria to evaluate the method performance, including image-based [
There is a parameter in our proposed method, region size (the size of superpixel), which impacts the performance of our proposed method. How to choose a suitable size becomes a critical problem in our experiment. Besides, for the same test sample, different classifiers may obtain the different classification results. So the choosing of classifier is equally important.
In our first experiment, two supervised classifiers, FDA and
Sensitivity versus
According to Figure
Besides, judging from Figure
In our second experiment, we will find the optimal region size based on pixel-based criterion [
ROC curves of pixel-based criterion on testing set with varying region sizes.
Judging from the above experiment results, it is easy to find that the performances of all the region sizes in the pixel-based criterion are relatively lower than those in the image-based criterion. The reason lies in the fact that the range of ground truth provided in database is not very precise and always larger than the true size of red lesions. So when each obtained red lesion map compares with its ground truth, the imprecise ground truth will affect the method’s performance. What is more, choosing the smaller size of superpixel, the segmentation results are more accurate than the bigger ones and when the region size is set to 10, our proposed algorithm achieves the best performance (i.e.,
Average execute time of per image and AUC with varying region sizes at pixel-based criterion.
AUC | Average time in seconds | |
---|---|---|
Region size = 10 | 0.74 | 116.23 s |
Region size = 30 | 0.73 | 38.05 s |
Region size = 50 | 0.70 | 10.37 s |
In addition, considering the clinical point of view and for screening applications [
In our last experiment, we employ image-based evaluation method proposed by Kauppi et al. [
Performance results on DiaretDB1 dataset.
Authors | Sensitivity | Specificity |
---|---|---|
Proposed method | 83.30% | 97.30% |
Sánchez et al. [ |
87.69% | 92.44% |
Ravishankar et al. [ |
95.10% | 90.50% |
Jaafar et al. [ |
98.80% | 86.20% |
Roychowdhury et al. [ |
75.50% | 93.73% |
From the comparison results of various algorithms illustrated in Table
To summarize, we put forward a novel red lesion detection based on superpixel Multichannel Multifeature (MCMF) classification in color retinal images, which is able to detect the red lesions efficiently regardless of their variability in appearance and size. Firstly, the whole image is segmented into a series of candidates using superpixel segmentation. And then, multiple features from the multichannel images as well as the contextual feature are proposed for describing each candidate. Next, FDA is introduced to classify the red lesions among the candidates. Finally, a postprocessing technique is applied to distinguish red lesions from blood vessels and fovea. Experiment results on DiaretDB1 database demonstrate that our proposed method is effective for red lesion detection.
Since our proposed approach extracts a number of features for each superpixel, complex relationships among the extracted features exist and other classifier (e.g., neural network or Extreme Learning Machine) may lead better classification result which could be validated and researched in future works. In addition, applying our proposed framework to other lesions detection is also another interesting topic for future study.
All authors declare that the support for this research does not lead to any conflicts of interest regarding the publication of this paper.
This work is supported by National Natural Science Foundation of China (nos. 61471110 and 61602221), Foundation of Liaoning Educational Department (L2014090), and Fundamental Research Funds for the Central Universities (N140403005, N162610004, and N160404003).