Considering the small differences between different types in the diabetic retinopathy (DR) grading task, a retinopathy grading algorithm based on cross-layer bilinear pooling is proposed. Firstly, the input image is cropped according to the Hough circle transform (HCT), and then the image contrast is improved by the preprocessing method; then the squeeze excitation group residual network (SEResNeXt) is used as the backbone of the model, and a cross-layer bilinear pooling module is introduced for classification. Finally, a random puzzle generator is introduced in the training process for progressive training, and the center loss (CL) and focal loss (FL) methods are used to further improve the effect of the final classification. The quadratic weighted Kappa (QWK) is 90.84% in the Indian Diabetic Retinopathy Image Dataset (IDRiD), and the area under the receiver operating characteristic curve (AUC) in the Messidor-2 dataset (Messidor-2) is 88.54%. Experiments show that the algorithm proposed in this paper has a certain application value in the field of diabetic retina grading.
In order to address the issues of spatial induction bias and lack of effective representation of global contextual information in colon polyp image segmentation, which lead to the loss of edge details and mis-segmentation of lesion areas, a colon polyp segmentation method that combines Transformer and cross-level phase-awareness is proposed. The method started from the perspective of global feature transformation, and used a hierarchical Transformer encoder to extract semantic information and spatial details of lesion areas layer by layer. Secondly, a phase-aware fusion module (PAFM) was designed to capture cross-level interaction information and effectively aggregate multi-scale contextual information. Thirdly, a position oriented functional module (POF) was designed to effectively integrate global and local feature information, fill in semantic gaps, and suppress background noise. Fourthly, a residual axis reverse attention module (RA-IA) was used to improve the network’s ability to recognize edge pixels. The proposed method was experimentally tested on public datasets CVC-ClinicDB, Kvasir, CVC-ColonDB, and EITS, with Dice similarity coefficients of 94.04%, 92.04%, 80.78%, and 76.80%, respectively, and mean intersection over union of 89.31%, 86.81%, 73.55%, and 69.10%, respectively. The simulation experimental results show that the proposed method can effectively segment colon polyp images, providing a new window for the diagnosis of colon polyps.
The existing retinal vessels segmentation algorithms have various problems that the end of main vessels are easy to break, and the central macula and the optic disc boundary are likely to be mistakenly segmented. To solve the above problems, a novel retinal vessels segmentation algorithm is proposed in this paper. The algorithm merged together vessels contour information and conditional generative adversarial nets. Firstly, non-uniform light removal and principal component analysis were used to process the fundus images. Therefore, it enhanced the contrast between the blood vessels and the background, and obtained the single-scale gray images with rich feature information. Secondly, the dense blocks integrated with the deep separable convolution with offset and squeeze-and-exception (SE) block were applied to the encoder and decoder to alleviate the gradient disappearance or explosion. Simultaneously, the network focused on the feature information of the learning target. Thirdly, the contour loss function was added to improve the identification ability of the blood vessels information and contour information of the network. Finally, experiments were carried out on the DRIVE and STARE datasets respectively. The value of area under the receiver operating characteristic reached 0.982 5 and 0.987 4, respectively, and the accuracy reached 0.967 7 and 0.975 6, respectively. Experimental results show that the algorithm can accurately distinguish contours and blood vessels, and reduce blood vessel rupture. The algorithm has certain application value in the diagnosis of clinical ophthalmic diseases.
Objective To investigate the effectiveness of repairing severe cicatricial contracture deformity in the web-space by kite-like incision combined with full-thickness skin grafting. Methods Between June 2008 and September 2011, 31 patients (87 web-spaces) with severe cicatricial contracture deformities in the web-spaces were treated. There were 24 males and 7 females, aged 5-43 years (median, 22 years). The causes of injuries were flame burn (26 cases), scald (3 cases), electric arc burn (1 case), and chemical burn (1 case). The degree of burn was deep second degree (14 cases) and third degree (17 cases). The interval time from injury to operation was 10 months to 17 years (median, 2.2 years). The kite-like incision was marked on the scar in the web-space. The rhombic scar between the adjacent metacarpophalangeal joints was excised, and cicatricial contracture was released completely. The secondary wound in the web-space was repaired with full-thickness autogeneic skin grafting. The secondary wound at donor site was directly sutured. Results All full-thickness skin grafts survived well. The incisions at donor sites healed primarily. Of 31 patients, 29 (82 web-spaces) were followed up 6-18 months (mean, 13 months). The sizes and depths of reconstructed web-spaces were similar to those of normal ones. No secondary cicatricial contracture was observed, and the function of fingers recovered well. Conclusion The short-term effectiveness is satisfactory by kite-like incision combined with full-thickness skin grafting for repairing severe cicatricial contracture deformities in the web-space, while the long-term effectiveness needs further observation.
Objective To investigate the feasibility of repairing secondary wounds after scar excision in patients with scar contracture in whole dorsum of the hands by expanded full-thickness skin grafts. Methods Between February 2008 and February 2011, 15 patients (19 hands) with scar contracture in whole dorsum of the hands were treated. There were 12 males (14 hands) and 3 females (5 hands), aged 18-43 years (mean, 28.6 years). The disease duration was 6-22 months (mean, 13.2 months). All cases complicated by different degrees of metacarpophalangeal joint extension deformity and interphalangeal joint flexion deformity. In one-stage operation, soft tissue expanders were implanted at donor sites, and the time of water injection was 4-5 months; in two-stage operation, scar was excised and secondary wound (14 cm × 9 cm-25 cm × 16 cm in size) was repaired with the expanded full-thickness skin grafts. Results After one-stage operation, the others had good blood supply except 1 case of skin necrosis in the center of expanded skin flap. After two-stage operation, the full-thickness skin grafts survived well in 18 hands; subepidermal effusion occurred in 1 hand, which was cured after changing dressing. The incisions at donor sites healed well in all cases. All patients were followed up 6-18 months (mean, 13.7 months). The grafts were excellent in color, texture, and elasticity, and no hypertrophic scar was observed at the edge and joint of the grafts. The hand function recovered well; the flexion and extension of wrist joint, metacarpophalangeal joint and interphalangeal joint were normal. Conclusion The expanded full-thickness skin graft is a suitable solution to repair effectively scar contracture in whole dorsum of the hands because it has good results in the cosmetic appearance and function recovery of the hands.
Objective To evaluate the cl inical effect and the pathological characteristics of acellular allogeneic dermal matrix in repairing unstable burn scar. Methods From January 2007 to June 2008, 19 cases of unstable burn scars (24 parts) were treated, including 16 males (20 parts) and 3 females (4 parts) with a median age of 27 years (range, 3-58 years). Theinjury was caused by flame (14 cases, 18 parts), electricity (4 cases, 5 parts), and hot water (1 case, 1 part). The unstable burn scars located on hands (8 cases), forearms (2 cases), thighs (3 cases), legs (2 cases), feet (2 cases), chest (1 case), and abdomen (1 case). Scar formed for 3 months to 1 year. The area of defect varied from 7 cm × 5 cm to 22 cm × 15 cm after scar removal. Defects were covered with acellular allogeneic dermal matrix and autogenous spl it-thickness skin graft. At 6-18 months after operation, the pathological observations of the epidermis, the basal membrane, and structural components of the dermis were done. Results All wounds healed by first intention. Scar ulcer disappeared completely in 18 cases and the composite skin grafts all survived. Some bl isters occurred in 1 case and were cured after dressing changing. All patients were followed up 10 months to 2 years (18 months on average). The grafted-skin was excellent in the appearance, texture, and elasticity. The function recovered well. Only superficial scar was observed at skin donor sites. Pathological observation showed that the epidermis and the basal membrane of the skin grafts were similar to that of normal skin, and no significant difference was found in newly capillaries between them. Collagen fibers arranged regularly, and there were few inflammatory cells in the matrix. Conclusion Acellular allogeneic dermal matrix with autogenous spl it-thickness skin graft may effectivly repair the wound after removing the unstable burn scar, and its structure is similar to that of normal skin.