Segmentation of Pigment Signs in Fundus Images for Retinitis Pigmentosa Analysis by Using Deep Learning

Frucci, Maria; Riccio, Daniel; Perna, Luigi Di; Simonelli, Francesca

doi:10.1007/978-3-030-30645-8_40

Cited by 6 publications

(7 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Table 5 presents the numerical results for retinal pigment sign detection based on G1, whereas Table 6 presents the same results based on G2. As shown in Table 5, most accuracies including Sen, Spe, F, and Acc by our method are higher than those by the state-of-the-art methods although P by our method is a little lower than that of the previous method [47]. In addition, our method outperforms the previous method [46] as shown in Table 6, and receiver operating characteristic (ROC) curves by our method are shown in Figure 6.…”

Section: Retinal Pigment Sign Segmentation Results By Rps-netmentioning

confidence: 67%

“…U-Net) [47] Subsequently improved segmentation performance by modified U-Net model, and 15% improvement in F-measure compared to [46].…”

Section: Brancati Et Al (Modifiedmentioning

confidence: 97%

“…As false negatives are more important in medical applications, so the Random Forest classifier shows more false negatives compared to AdaBoost [46]. The same group subsequently enhanced the accuracy further by using a modified U-Net deep learning model on patches [47]. Therefore, the two blocks removed from the original five U-Net blocks and the number of filters is halved.…”

Section: Retinal Image Segmentation Based On Deep-feature (Cnn)mentioning

confidence: 99%

“…Because this study is based on the rare retinitis pigmentosa disease with applications for the method of RP analysis to aid the medical practitioner in early diagnosis of the disease, we used RIPS dataset which is the only publicly available real dataset [46]. The same RIPS dataset was solely used by previous studies [33,46,47]. Moreover, to validate the performance of RPS-Net, 4-fold cross-validation is used with different patients for training and testing.…”

Section: Experimental Data and Environmentmentioning

confidence: 99%

“…Avoiding the preprocessing schemes and a substantial increase in F-measure compared to the machine learning method are the strength of this method. Considering the F-measure, this method improved overall segmentation performance, but as false negatives (represented by sensitivity) are considered more critical than false positive, this method [ 47 ] has higher false negatives (low sensitivity) compared to Random Forest and AdaBoost presented in [ 46 ].…”

Section: Related Workmentioning

confidence: 99%

See 4 more Smart Citations

Deep Learning-Based Detection of Pigment Signs for Analysis and Diagnosis of Retinitis Pigmentosa

Arsalan

Baek

Owais

et al. 2020

Sensors

View full text Add to dashboard Cite

Ophthalmological analysis plays a vital role in the diagnosis of various eye diseases, such as glaucoma, retinitis pigmentosa (RP), and diabetic and hypertensive retinopathy. RP is a genetic retinal disorder that leads to progressive vision degeneration and initially causes night blindness. Currently, the most commonly applied method for diagnosing retinal diseases is optical coherence tomography (OCT)-based disease analysis. In contrast, fundus imaging-based disease diagnosis is considered a low-cost diagnostic solution for retinal diseases. This study focuses on the detection of RP from the fundus image, which is a crucial task because of the low quality of fundus images and non-cooperative image acquisition conditions. Automatic detection of pigment signs in fundus images can help ophthalmologists and medical practitioners in diagnosing and analyzing RP disorders. To accurately segment pigment signs for diagnostic purposes, we present an automatic RP segmentation network (RPS-Net), which is a specifically designed deep learning-based semantic segmentation network to accurately detect and segment the pigment signs with fewer trainable parameters. Compared with the conventional deep learning methods, the proposed method applies a feature enhancement policy through multiple dense connections between the convolutional layers, which enables the network to discriminate between normal and diseased eyes, and accurately segment the diseased area from the background. Because pigment spots can be very small and consist of very few pixels, the RPS-Net provides fine segmentation, even in the case of degraded images, by importing high-frequency information from the preceding layers through concatenation inside and outside the encoder-decoder. To evaluate the proposed RPS-Net, experiments were performed based on 4-fold cross-validation using the publicly available Retinal Images for Pigment Signs (RIPS) dataset for detection and segmentation of retinal pigments. Experimental results show that RPS-Net achieved superior segmentation performance for RP diagnosis, compared with the state-of-the-art methods.

show abstract

Section: Retinal Pigment Sign Segmentation Results By Rps-netmentioning

confidence: 67%

“…U-Net) [47] Subsequently improved segmentation performance by modified U-Net model, and 15% improvement in F-measure compared to [46].…”

Section: Brancati Et Al (Modifiedmentioning

confidence: 97%