One of the most leading causes of death worldwide is Colorectal cancer(CRC). Polyp segmentation is the most important detected measure for preventing CRC. However, there is still a missing rate for diminutive polyps and multiple ones. In order to solve the phenomenon, we propose to introduce auxiliary attention module(AAM) that can enhance the learning of features related to multiple and diminutive polyps by focusing more on the located and detailed information. Meanwhile, we design to decrease missed rate of multiple and diminutive polyps by implementing an area adaptive loss(AAL) which adapts the weight according to the area and the number of polyps. Our proposed novel AAM and AAL concentrates on training with hard examples and localized information. To evaluate the effectiveness and generalization ability of our proposed model, We utilize three different datasets of variable sizes and a cross dataset. Our proposed method achieves the best results on the Kvasir-SEG dataset, the CVC-ClinicDB dataset and the cross dataset, particularly for the Kvasir-Sessile dataset consisting of small,flat and diminutive polyps. Extensive experimental results show that our proposed DoubleAANet surpass the performance of all existing state-of-the-art segmentation methods.

In the realm of computer vision, semantic segmentation is the task of recognizing objects in images at the pixel level. This is done by performing a classification of each pixel. The task is complex and requires sophisticated skills and knowledge about the context to identify objects’ boundaries. The importance of semantic segmentation in many domains is undisputed. In medical diagnostics, it simplifies the early detection of pathologies, thus mitigating the possible consequences. In this work, we provide a review of the literature on deep ensemble learning models for polyp segmentation and we develop new ensembles based on convolutional neural networks and transformers. The development of an effective ensemble entails ensuring diversity between its components. To this end, we combine different models (HarDNet-MSEG, Polyp-PVT, and HSNet) trained with different data augmentation techniques, optimization methods, and learning rates, which we experimentally demonstrate to be useful to form a better ensemble. Most importantly, we introduce a new method to obtain the segmentation mask which is more suitable for combining transformers in an ensemble. In our extensive experimental evaluation, the proposed ensembles exhibit state-of-the-art performance.

Ensuring precise segmentation of colorectal polyps holds critical importance in the early diagnosis and treatment of colorectal cancer. Nevertheless, existing deep learning-based segmentation methods are fully-supervised, requiring extensive precise manual pixel-level annotation data, which leads to high annotation costs. Additionally, it remains challenging to train large-scale segmentation models when confronted with limited colonoscopy data. To address these issues, we introduce the general segmentation foundation model—Segment Anything Model (SAM) into the field of medical image segmentation. Fine-tuning the foundation model is an effective approach to tackle sample scarcity. However, current SAM fine-tuning techniques still rely on precise annotations. To overcome this limitation, we propose WSPoly-SAM, a novel weakly-supervised approach for colonoscopy polyp segmentation. WSPoly-SAM utilizes weak annotations to guide SAM in generating segmentation masks, which are then treated as pseudo-labels to guide the fine-tuning of SAM, thereby reducing the dependence on precise annotations data. To improve the reliability and accuracy of pseudo-labels, we have designed a series of enhancement strategies to improve the quality of pseudo-labels and mitigate the negative impact of low-quality pseudo-labels. Experimental results on five medical image datasets demonstrate that WSPoly-SAM outperforms current fully-supervised mainstream polyp segmentation networks on the Kvasir-SEG, ColonDB, CVC-300, and ETIS datasets. Furthermore, by using different amounts of training data in weakly-supervised and fully-supervised experiments, it is found that weakly-supervised fine-tuning can save 70% to 73% of annotation time costs compared to fully-supervised fine-tuning. This study provides a new perspective on the combination of weakly-supervised learning and SAM models, significantly reducing annotation time and offering insights for further development in the field of colonoscopy polyp segmentation.

The advent of large models, also known as foundation models, has significantly transformed the AI research landscape, with models like Segment Anything (SAM) achieving notable success in diverse image segmentation scenarios. Despite its advancements, SAM encountered limitations in handling some complex low-level segmentation tasks like camouflaged object and medical imaging. In response, in 2023, we introduced SAM-Adapter, which demonstrated improved performance on these challenging tasks. Now, with the release of Segment Anything 2 (SAM2)—a successor with enhanced architecture and a larger training corpus—we reassess these challenges. This paper introduces SAM2-Adapter, the first adapter designed to overcome the persistent limitations observed in SAM2 and achieve new state-of-the-art (SOTA) results in specific downstream tasks including medical image segmentation, camouflaged (concealed) object detection, and shadow detection. SAM2-Adapter builds on the SAM-Adapter's strengths, offering enhanced generalizability and composability for diverse applications. We present extensive experimental results demonstrating SAM2-Adapter's effectiveness. We show the potential and encourage the research community to leverage the SAM2 model with our SAM2-Adapter for achieving superior segmentation outcomes. Code, pre-trained models, and data processing protocols are available at http://tianrun-chen.github.io/SAM-Adaptor/.

Current sampling methods for detecting the presence of the invasive freshwater hydrozoan Craspedacusta sowerbii rely mainly on visual confirmation of the medusa stage. Confirming the presence of the polyp stage is equally important to observing medusae since typical late summer/early fall occurrences or observations of medusae are sporadic though becoming more frequent. The polyp stage is important as it is the organism’s primary stage and is present throughout the year depending on water temperatures. Therefore, sampling methods for the polyp stage are commonly the collection of substrates such as rocks, plants or pieces of wood in a given body of water and can be cumbersome to examine. Based on preliminary culturing of the polyp stage on glass and plastic microscope slides in the lab, we designed a novel sampling methodology based on submerging 4 substrate types (small glass petri dishes, Hester-Dendy plates, plastic and glass microscope slides) to confirm the presence of C. sowerbii polyps in the field. We tested this method in various lakes in the Illinois-Indiana region (USA). The sampling method we designed was effective in that C. sowerbii polyps (and sometimes colonies) were found on both plastic and glass slides. While this method can be sufficient for detection of the polyp stage, it also shows significant potential for improvement; we highlight water flow and competition and predation from other organisms as significant factors influencing the collection of C. sowerbii polyps for future methodologies.

Hereditary polyposis syndromes are rare but potentially devastating conditions which require multidisciplinary care from an early age. Societal guidelines serve as the framework for disease diagnosis, surveillance and treatment. Guidelines are based on expert opinion and ideally are evidence-based; however, recommendations may vary among different societies and can result in discordant care. This paper aims to summarize key differences in management based on societal guidelines for pediatric polyposis syndromes and identify some of the factors which may contribute to divergence in care.

Primary lymphoma of the middle ear is rare in cats, and little information has been accumulated on the treatment and course. An 11-year-old spayed female mixed-breed cat was brought to our hospital with facial nerve paralysis, which had been stable using prednisolone for 3 months. She was diagnosed with B-cell lymphoma of the right middle ear after otoscopic sampling, which showed evidence of filling of bilateral tympanic bullae on computed tomography. Hepatic metastasis was suspected and she was treated with multidrug chemotherapy, resulting in clinical remission. On day 176, magnetic resonance imaging and computed tomography was performed as a checkup, revealing tumor invasion into the nasopharyngeal region and recurrence of hepatic lesions without any brain abnormality. Nasal congestion then worsened and the patient died on day 228 after presenting with generalized seizures. Clinicians should be aware of the possibility of invasion to surrounding tissues and hepatic metastasis in cases of feline primary middle ear lymphoma.

Chronic rhinosinusitis is a common inflammatory disease of paranasal sinuses, which causes rhinorrhea, nasal congestion and hyposmia. The genetic predisposition or the exposure to irritants can sustain the inflammatory response and the development of nasal polyposis. Nasal polyps are benign and teardrop-shaped growths that project in the nasal cavities and originate from the ethmoid sinuses. This inflammatory process is associated with high expression of IL-5 cytokine and infiltration of eosinophils. Humanized monoclonal antibodies targeting IL-5 or its receptor, represent a therapeutic strategy in the treatment of nasal polyposis in combination with corticosteroids. The molecular pathogenesis of nasal polyps in CRS patients is associated to the epithelial-mesenchymal transition (EMT), a process in which epithelial cells lose their typical phenotype acquiring a mesenchymal phenotype. TGFβ/SMAD, ERK, and Wnt/β-catenin pathways are altered in EMT during the nasal tissue remodeling. miRNA and inhibitor molecules targeting these altered signaling pathways are able to interfere with EMT; which could lead to alternative therapies. Nasal polyps are an alternative source of mesenchymal stem cells which can be easily isolated from surgical biopsies. A molecular understanding of the biology of PO-MSCs will contribute to delineating inflammatory process underlying the development of nasal polyps.

Objectives: We aimed to clarify the endoscopic and clinicopathological features of raspberry-shaped gastric lesions (RSGLs) and to establish an endoscopic diagnostic algorithm for RSGLs. Methods: We collected RSGLs from an endoscopic database at our hospital between May 2009 and August 2021. All RSGLs were histopathologically classified and compared based on their endoscopic and clinicopathological characteristics. Results: Sixty-five RSGLs in 54 patients were classified into five histopathological types: gastric adenocarcinoma of foveolar type (GA-FV, n = 43), gastric adenocarcinoma of fundic-gland type (GA-FG, n = 2), gastric adenocarcinoma of fundic-gland mucosa type (GA-FGM, n = 4), hyperplastic polyp (HP, n = 12), and proton pump inhibitor-related lesion (PPI-L, n = 4). All RSGLs exhibited polygonal or curved marginal crypt epithelium (MCE). GA-FV lesions had homogenously reddish (95%) and irregular microvascular (MV) pattern (91%). GA-FG lesions were heterogeneously reddish with a submucosal tumor shape (100%) and had a regular MV pattern (50%). GA-FGM lesions were homogenously reddish (75%) and occasionally had a submucosal tumor shape (50%) with an irregular MV pattern (75%). HP and PPI-L were homogeneously reddish (93%), with linear or dotted MCE (81%) and a regular MV pattern (100%). Conclusion: Our diagnostic algorithm for RSGLs constructed using endoscopic features might be useful for the endoscopic differential diagnosis of RSGLs.