Areas Of Pathology Research Articles

Detection and Removal of Assymmetrical Skin Lesions Using DU-Net for Patch Extraction

The incidence of melanoma, a highly fatal kind of malignant tumor, is exhibiting an upward trend in its progression. Timely intervention and management of skin cancer significantly enhance the likelihood of survival. Analyzing dermoscopic images poses challenges for dermatologists due to various distracting factors, including fluctuations in lighting and reflections on the skin's surface. Accurately delineating the area of skin pathology is crucial for diagnosing the specific skin condition. This study introduces DU-net, a method for identifying and removing pigmented skin lesions. The DU-net framework incorporates deep convolutional neural networks, notably YOLOv5, to perform patch detection. It also employs asymmetrical patch contouring methods to preserve edge information. Additionally, clustering algorithms are utilized to identify pixel groups and extract patches from the image. The De Trop Noise Exclusion technique, in conjunction with the process of in-painting, is employed to successfully eliminate hair from the images within the ISIC-2018 and 2019 datasets. Rigorous annotation of skin images with lesions of various sizes and shapes using rectangle bounding is carried out, and YOLOv5 hyper-parameters are fine-tuned to achieve high-confidence multiple lesion detection in dermoscopic images. Despite complex textures and unclear boundaries, our approach consistently detects and labels patches, accurately segmenting the areas of skin pathology. The model's performance is assessed on these datasets using various parameter metrics, the results of the study indicate that the segmentation strategies described in this research exhibit an average accuracy ranging from about 92% to 94%.

Homozygous CNP Mutation and Neurodegeneration in Weimaraners: Myelin Abnormalities and Accumulation of Lipofuscin-like Inclusions.

A progressive neurological disorder was observed in a male neutered Weimaraner. Clinical signs included fecal incontinence, lethargy, moderate paraparesis, proprioceptive pelvic limb ataxia, falling, cognitive decline, incoordination, decreased interest in food, changes in posture, and episodes of trance-like behavior. Neurologic signs were first observed at approximately 4 years, 10 months of age and progressed slowly. Magnetic resonance imaging showed generalized brain atrophy with areas of white matter pathology. Humane euthanasia was elected at 6 years, 7 months of age due to increasing severity of the neurological signs. Autofluorescent intracellular granules were observed in the cerebral and cerebellar cortexes, optic nerve, and cardiac muscle of the affected dog. These abnormal inclusions in the cerebral cortex and cardiac muscle immunolabeled with antibodies to mitochondrial ATP synthase subunit c protein, like that observed in the neuronal ceroid lipofuscinosis group of lysosomal storage diseases. Immunolabeling also demonstrated pronounced neuroinflammation in brain tissues. The ultrastructural appearances of the disease-related inclusion bodies in the brain and optic nerve were quite variable. The ultrastructure and locations of many of the inclusions in the nervous tissues suggested that they were derived, at least in part, from the myelin surrounding axons. The storage bodies in the cardiac muscle were located in mitochondria-rich regions and consisted of parallel arrays of membrane-like components interspersed with electron-dense flocculent material. The disease was characterized by pronounced abnormalities in the myelin of the brain and optic nerve consisting of distinctive areas of ballooning between the layers of myelin. The whole genome sequence generated from the affected dog contained a homozygous G-to-A missense mutation in CNP, which encodes proteins with CNPase enzyme activity and a structural role in myelin. The mutation predicts a Thr42Met amino acid sequence substitution. Genotyping of archived Weimaraner DNA samples identified an additional G > A variant homozygote with a clinical history and brain lesions similar to those of the proband. Of 304 Weimaraners and over 4000 other dogs of various breeds, the proband and the other Weimaraner that exhibited similar signs were the only two that were homozygous for the CNP missense variant. CNPase immunolabeling was widespread in brain tissues from normal dogs but was undetectable in the same tissues from the proband. Based on the clinical history, fluorescence and electron-microscopy, immunohistochemistry, and molecular genetic findings, the late-onset Weimaraner disorder likely results from the missense mutation that results in CNPase deficiency, leading to myelin abnormalities, accumulation of lysosomal storage bodies, and brain atrophy. Similar disorders have been associated with different CNP variants in Dalmatians and in human subjects.

Investigation of How Corneal Densitometry Artefacts Affect the Imaging of Normal and Keratoconic Corneas.

To investigate corneal densitometry artefacts found in Pentacam Scheimpflug scans and their potential effect on assessing keratoconic (KC) corneas compared to normal (N) corneas. The current study utilises Pentacam data of 458 N eyes, aged 35.6 ± 15.8 (range 10-87), referred to as the "N group", and 314 KC eyes, aged 31.6 ± 10.8 (range 10-72), referred to as the "KC group", where densitometry data were extracted and analysed via a custom-built MATLAB code. Radial summations of the densitometry were calculated at diameters ranging from 0.5 mm to 5.0 mm. The minimum normalised radial summation of densitometry (NRSD) value and angle were determined at each diameter and then linked. KC cone locations and areas of pathology were determined, and a comparison between N and KC groups was carried out both within the averaged area of pathology and over the corneal surface. Joining minimum NRSD trajectory points marked a clear distortion line pointing to the nasal-superior direction at 65° from the nasal meridian. The findings were found to be independent of eye laterality or ocular condition. Consistency was detected in the right and left eyes among both the N and KC groups. The location of the KC cone centre and the area of pathology were determined, and the densitometry output was compared both within the area of pathology and over the whole cornea. When the average densitometry was compared between N and KC eyes within the KC area of pathology, the N group recorded a 16.37 ± 3.15 normalised grey-scale unit (NGSU), and the KC group recorded 17.74 ± 3.4 NGSU (p = 0.0001). However, when the whole cornea was considered, the N group recorded 16.71 ± 5.5 NGSU, and the KC group recorded 15.72 ± 3.98 NGSU (p = 0.0467). A weak correlation was found between the Bad D index and NGSU when the whole measured cornea was considered (R = -0.01); however, a better correlation was recorded within the KC area of pathology (R = 0.21). Nasal-superior artefacts are observed in the densitometry Pentacam maps, and analysis shows no significant differences in their appearance between N or KC corneas. When analysing KC corneas, it was found that the cone positions are mostly on the temporal-inferior side of the cornea, opposite to the densitometry artefact NRSD trajectory. The analysis suggests that the corneal densitometry artefacts do not interfere with the KC area of pathology as it reaches its extreme in the opposite direction; therefore, weighting the densitometry map to increase the contribution of the inferior-temporal cornea and decreasing that of the superior-nasal area would improve the classification or identification of KC if densitometry is to be used as a KC metric.

Global Publication Trends and Research Hotspots of The Gastric Neuroendocrine Neoplasms: A Bibliometric Analysis of The Current Situation

Objective: Gastric neuroendocrine neoplasms (gNENs) are uncommon tumors, with growing understandings about the disease. Bibliometric analyzes have the advantage of visually depicting the dynamic evaluation of scientific knowledge of a specific topic. The aim of this study was to perform and report bibliometric analysis of gNENs, which was not formerly studied in the literature. Methods: Articles published between 1980 and 2022 within the database of Web of Science Core Collection were included in this bibliometric analysis. Vosviewer package program and Datawrapper were used for bibliometric data interpretation. Results: A total of 2270 articles about gNENs was detected with 63240 citations and an H index of 103. A remarkable increase was detected among the articles for years. Research have focused on gastroenterology and hepatology, endocrinology and metabolism, oncology, general medicine, pathology and surgery areas. Norwegian University of Science Technology was the leading institution about gNENs literature. Modlin IM, had the highest number of articles and citations among the authors. United States, Japan and Italy were the top three countries with the most published articles. Conclusion: This bibliometric study provides an engrossing, insightful conclusion to the research and development trajectory in gNENs with a future perspective.

Non-invasive cardiac activation mapping and identification of severity of epicardial substrate in Brugada Syndrome: a case report

IntroductionIt has recently been shown that electrocardiographic imaging (ECGi) can be employed in individuals undergoing an ajmaline test who have Brugada Syndrome (BrS), to evaluate the extent of substrate-involved arrhythmia in the right ventricular overflow tract (RVOT). For the first time, we stratify the risk of sudden cardiac death (SCD) in BrS during ajmaline testing using the dST-Tiso interval (a robust predictor of the inducibility of ventricular arrhythmias (VAs) in the presence of drug-induced BrS type-1 pattern) in combination with ECGi technology.Case presentationWe studied a 48-year-old man with BrS ECG type-2 pattern and presence of J-wave without a family history of SCD but with a previous syncope. Transthoracic echocardiography and cardiac magnetic resonance imaging were performed, showing normal results. The ECG was performed to assess the novel ECG marker “dST-Tiso interval.” The 3D epicardial mapping of the RVOT surface was performed with the support of a non-contact cardiac mapping system in sinus rhythm during ajmaline infusion. The examination of the propagation map unveiled the presence of multiple conduction blocks in this pathologic epicardial region, and the conduction blocks were identified within the central part and/or near the boundary separating the normal and slow conduction areas.ConclusionThe dST-Tiso interval, which lies between the onset and termination of the coved ST-segment elevation and serves as a robust predictor of VA inducibility in cases of drug-induced BrS type-1 pattern, was utilized in conjunction with ECGi technology (employed for the non-invasive confirmation and identification of the pathological substrate area). This combined approach was applied to stratify the risk of SCD in BrS during ajmaline testing, alongside clinical scores.

Improving early detection of keratoconus by Non Contact Tonometry. A computational study and new biomarkers proposal

Keratoconus is a progressive ocular disorder affecting the corneal tissue, leading to irregular astigmatism and decreased visual acuity. The architectural organization of corneal tissue is altered in keratoconus, however, data from ex vivo testing of biomechanical properties of keratoconic corneas are limited and it is unclear how their results relate to true mechanical properties in vivo. This study explores the mechanical properties of keratoconic corneas through numerical simulations of non-contact tonometry (NCT) reproducing the clinical test of the Corvis ST device. Three sensitivity analyses were conducted to assess the impact of corneal material properties, size, and location of the pathological area on NCT results. Additionally, novel asymmetry-based indices were proposed to better characterize corneal deformations and improve the diagnosis of keratoconus. Our results show that the weakening of corneal material properties leads to increased deformation amplitude and altered biomechanical response. Furthermore, asymmetry indices offer valuable information for locating the pathological tissue. These findings suggest that adjusting the Corvis ST operation, such as a camera rotation, could enhance keratoconus detection and provide insights into the relative position of the affected area. Future research could explore the application of these indices in detecting early-stage keratoconus and assessing the fellow eye’s risk for developing the pathology.

Applications of artificial intelligence in the field of oral and maxillofacial pathology: a systematic review and meta-analysis

BackgroundSince AI algorithms can analyze patient data, medical records, and imaging results to suggest treatment plans and predict outcomes, they have the potential to support pathologists and clinicians in the diagnosis and treatment of oral and maxillofacial pathologies, just like every other area of life in which it is being used. The goal of the current study was to examine all of the trends being investigated in the area of oral and maxillofacial pathology where AI has been possibly involved in helping practitioners.MethodsWe started by defining the important terms in our investigation's subject matter. Following that, relevant databases like PubMed, Scopus, and Web of Science were searched using keywords and synonyms for each concept, such as "machine learning," "diagnosis," "treatment planning," "image analysis," "predictive modelling," and "patient monitoring." For more papers and sources, Google Scholar was also used.ResultsThe majority of the 9 studies that were chosen were on how AI can be utilized to diagnose malignant tumors of the oral cavity. AI was especially helpful in creating prediction models that aided pathologists and clinicians in foreseeing the development of oral and maxillofacial pathology in specific patients. Additionally, predictive models accurately identified patients who have a high risk of developing oral cancer as well as the likelihood of the disease returning after treatment.ConclusionsIn the field of oral and maxillofacial pathology, AI has the potential to enhance diagnostic precision, personalize care, and ultimately improve patient outcomes. The development and application of AI in healthcare, however, necessitates careful consideration of ethical, legal, and regulatory challenges. Additionally, because AI is still a relatively new technology, caution must be taken when applying it to this industry.

Biobanco orientado nas doenças veterinárias para a análise biomolecular com base em tecidos congelados, cultura celular e tecido forense

ABSTRACT: Collecting and analyzing biological specimens leads to unprecedented opportunities for assessing and understanding the complex networks of interactions between biomolecules, and the functional and clinical consequences of their alterations. In this context, the Tissue Bank plays a key role in life science research and development, making its implementation in veterinary medicine essential for developing new research, especially in the forensic and tumor fields. In total, 52 tissues were collected, 15 forensic samples, 25 neoplasms, and 12 cell cultures. The storage, freezing, data management protocols, and the quality of these samples were analyzed. In the same way, the quality of the RNA and DNA in the short and long term was evaluated. The quality of the genetic material was confirmed by polymerase chain reaction. For data comparisons, non-parametric tests were used, with a significance level of p˂0.05. The measures adopted in the Tissue Bank for obtaining and handling the samples and for controlling and guaranteeing the quality were considered adequate, as they enabled optimal preservation of the integrity of the genetic material, a relevant factor for the use of tumor samples and forensic samples stored in the Biobank, thus contributing to future molecular research in the area of pathology and investigations of crimes against fauna.

Levels of evidence and grades of recommendation supporting European society for medical oncology clinical practice guidelines.

The European Society for Medical Oncology (ESMO) guidelines are among the most comprehensive and widely used clinical practice guidelines (CPGs) globally. However, the level of scientific evidence supporting ESMO CPG recommendations has not been systematically investigated. This study assessed ESMO CPG levels of evidence (LOE) and grades of recommendations (GOR), as well as their trends over time across various cancer settings. We manually extracted every recommendation with the Infectious Diseases Society of America (IDSA) classification from each CPG. We examined the distribution of LOE and GOR in all available ESMO CPG guidelines across different topics and cancer types. Among the 1,823 recommendations in the current CPG, 30% were classified as LOE I, and 43% were classified as GOR A. Overall, there was a slight decrease in LOE I (-2%) and an increase in the proportion of GOR A (+1%) in the current CPG compared to previous versions. The proportion of GOR A recommendations based on higher levels of evidence such as randomized trials (LOE I-II) shows a decrease (71% vs. 63%, p = 0.009) while recommendations based on lower levels of evidence (LOE III-V) show an increase (29% vs. 37%, p = 0.01) between previous and current version. In the current versions, the highest proportion of LOE I (42%) was found in recommendations related to pharmacotherapy, while the highest proportion of GOR A recommendations was found in the areas of pathology (50%) and diagnostic (50%) recommendations. Significant variability in LOE I and GOR A recommendations and their changes over time was observed across different cancer types. One-third of the current ESMO CPG recommendations are supported by the highest level of evidence. More well-designed randomized clinical trials are needed to increase the proportion of LOE I and GOR A recommendations, ultimately leading to improved outcomes for cancer patients.

Non-neoplastic B-cell predominant lymphoid proliferations at the organs exposed to external environment mimicking lymphoma: A potential diagnostic pitfall.

Background: Typically, lymphatic tissue proliferative lesions include either benign lesions or lymphoma. However, not all lymphatic lesions can currently be accurately classified into one category, particularly in mucosal areas that are in contact with the external environment.Aims: To explore the morphology, immunophenotype, and molecular changes of Non-neoplastic B-cell predominant lymphoid proliferations (NBPLP) in pathological areas that are exposed to external surroundings which mimicked lymphoma.Methods and Results: 18 cases of Atypical lymphoid hyperplasia (AtLP) were retrieved in this study. The biopsy samples were mucosal samples obtained from areas exposed to external surroundings, including intestines, urethra, cervix, tonsils, and tongue. Microscopically, there is a different level of B cell hyperplasia accompanied by morphological atypia. We categorized the morphology into 4 groups: type A (7/18), type B (3/18), type C (3/18), type D (5/18). Part of the AtLP was found positive for BCR gene rearrangement (6/15), and TCR gene rearrangement (1/4). The follow-up period ranged from 14.2 to 70 months. No evidence of lymphoma was found. Therefore, we diagnosed all of the presented cases as NBPLP. We illustrated the key differential points and provided valuable diagnostic experience on each subtype.Conclusions: Areas exposed to the external environment are commonly exposed to antigen and easily present with AtLP of NBPLP, accompanying with positive IGH rearrangement. Therefore, a comprehensive evaluation of macroscopic, morphology, immunophenotype, and molecular diagnostics is required to prevent the overdiagnosis of lymphoma.

Progressive brain structural abnormality in cerebral small vessel disease assessed with MR imaging by using causal network analysis

AimsCerebral small vessel disease (CSVD) is a complex condition characterized by a combination of microcirculation disorders and neurodegenerative processes, CSVD is associated with structural abnormalities in multiple brain regions. However, the progressive pattern of structural changes remains unknown. MethodsIn order to detail the progressive structural changes in CSVD patients according to the degree of cognitive impairment, we recruited 121 CSVD patients and 104 healthy controls (HCs). Voxel-based morphometry was employed to measure the gray matter volume (GMV) of each participant. According to the VICCCS-2 diagnostic criteria, patients were initially divided into three stage groups, then we investigated the GMV changes in each stage and their causal relationships using causal structure covariance network (CaSCN) analysis. ResultsOverall, patients with CSVD presented stage-specific GMV alterations compared with HCs. With the worsening of cognitive impairment, the decrease in gray matter volume starts from the right hippocampus and gradually spreads to the cortical-subcortical brain regions. Importantly, the right hippocampus in CSVD patients plays a driving role in the directional network and forms both positive and negative causal effect networks with cortical-subcortical brain regions. ConclusionsThis study reveals the significance of the right hippocampus as an early pathological area in CSVD patients and its causal impact on brain GMV changes with disease progression, shedding light on structural brain damage hierarchy and compensatory mechanisms.

Study of the kinetics of accumulation and distribution of the photosensitizer photoditazin in the oral mucosa in patients with lichen planus

Was to explore the accumulation and distribution of the photosensitizer Photoditazine in the oral mucosa when applied to pathological lesions in patients with severe forms of lichen planus. A clinical and laboratory examination was carried out in 50 patients with severe forms of lichen planus (bullous and erosive-ulcerative) aged 18 to 70 years, including 6 men and 44 women. For autofluorescent imaging a LED device with a wavelength in the violet region of the spectrum (400±10 nm) was used. Quantitative registration of the kinetics of accumulation and distribution of the photosensitizer was carried out using the method of local fluorescence spectroscopy by measuring the fluorescence spectra. The measurements were made before applying the photosensitizer, 10, 20 and 30 minutes after application. The study showed that in most patients with erosive-ulcerative and bullous forms of lichen planus, the accumulation of the photosensitizer in the lesions on the oral mucosa increased as the exposure time increased from 20 to 30 minutes. The fastest accumulation of the photosensitizer occurred in the areas of mucosal lesions with the most pronounced vascularization, namely, in the area of the tongue and the bottom of the oral cavity. Using the method of local fluorescence spectroscopy, the kinetics of accumulation and destruction of photosensitizer in pathological areas of the oral mucosa was determined, and therefore the optimal time of laser exposure to the lesion was determined.

Cervical spinal cord susceptibility-weighted MRI at 7T: Application to multiple sclerosis

BackgroundSusceptibility-weighted imaging (SWI) has been extensively studied in the brain and in diseases of the central nervous system such as multiple sclerosis (MS) providing unique opportunities to visualize cerebral vasculature and disease-related pathology, including the central vein sign (CVS) and paramagnetic rim lesions (PRLs). However, similar studies evaluating SWI in the spinal cord of patients with MS remain severely limited. PurposeBased on our previous findings of enlarged spinal vessels in MS compared to healthy controls (HCs), we developed high-field SWI acquisition and processing methods for the cervical spinal cord with application in people with MS (pwMS) and HCs. Here, we demonstrate the vascular variability between the two cohorts and unique MS lesion features in the cervical cord. MethodsIn this retrospective, exploratory pilot study conducted between March 2021 and March 2022, we scanned 12 HCs and 9 pwMS using an optimized non-contrast 2D T2*-weighted gradient echo sequence at 7 tesla. The overall appearance of the white and gray matter as well as tissue vasculature were compared between the two cohorts and areas of MS pathology in the patient group were assessed using both the magnitude and processed SWI images. ResultsWe show improved visibility of vessels and more pronounced gray and white matter contrast in the MS group compared to HCs, hypointensities surrounding the cord in the MS cohort, and identify signal changes indicative of the CVS and paramagnetic rims in 66 % of pwMS with cervical spinal lesions. ConclusionIn this first study of SWI at 7T in the human spinal cord, SWI holds promise in advancing our understanding of disease processes in the cervical cord in MS.

Reconstruction of the ulnar artery in Guyon’s canal with an arterial graft: Anatomical study

Ulnar artery thrombosis in Guyon’s canal can lead to vascular insufficiency in the fingers. The recommended treatment is resection and reconstruction of the pathological area. A bypass may be necessary, which may be venous or arterial. Arterial bypasses have better long-term patency; however, they are a source of donor-site complications.We carried out an anatomical study on 11 upper limbs and 7 lower limbs from cadavers to identify a technically accessible arterial graft, of a diameter suitable for bypassing the ulnar artery in Guyon’s canal and with acceptable scar sequelae (few predicted postoperative complications, discreet size and/or location of scar).Three grafts were considered: anterior interosseous artery, radial recurrent artery and descending genicular artery. The various grafts were dissected and harvested from cadaver specimens, then their lengths and diameters were measured. The diameter of the candidate grafts was compared to the diameter of the distal ulnar artery.The diameter of the descending genicular artery matched the ulnar artery better than the radial recurrent artery or the anterior interosseous artery (103% vs 44% and 67%, respectively). Mean graft length was 6.6 cm. The anatomical configuration of the descending genicular artery allowed Y-shaped bypasses to be performed. Harvesting this artery appears to cause little damage and allows bypasses up to 6 cm to be performed.Despite its smaller diameter making it necessary to perform a microvascular size adjustment, the anterior interosseous artery is a candidate graft because it is long enough (119 mm) and located near the surgical site. Level of evidenceV.

LGI Net: Enhancing local-global information interaction for medical image segmentation

Medical image segmentation is a critical task used to accurately extract regions of interest and pathological areas from medical images. In recent years, significant progress has been made in the field of medical image segmentation using deep learning and neural networks. However, existing methods still have limitations in terms of fusing local features and global contextual information due to the complex variations and irregular shapes of medical images. To address this issue, this paper proposes a medical image segmentation architecture called LGI Net, which improves the internal computation to achieve sufficient interaction between local perceptual capabilities and global contextual information within the network. Furthermore, the network incorporates an ECA module to effectively capture the interplay between channels and improve inter-layer information exchange capabilities. We conducted extensive experiments on three public medical image datasets: Kvasir, ISIC, and X-ray to validate the effectiveness of the proposed method. Ablation studies demonstrated the effectiveness of our LGAF, and comparative experiments confirmed the superiority of our proposed LGI Net in terms of accuracy and parameter efficiency. This study provides an innovative approach in the field of medical image segmentation, offering valuable insights for further improvements in accuracy and performance. The code and models will be available at https://github.com/LiuLinjie0310/LGI-Net.

Gene Expression Within a Human Choroidal Neovascular Membrane Using Spatial Transcriptomics.

Macular neovascularization is a relatively common and potentially visually devastating complication of age-related macular degeneration. In macular neovascularization, pathologic angiogenesis can originate from either the choroid or the retina, but we have limited understanding of how different cell types become dysregulated in this dynamic process. To study how gene expression is altered in focal areas of pathology, we performed spatial RNA sequencing on a human donor eye with macular neovascularization as well as a healthy control donor. We performed differential expression to identify genes enriched within the area of macular neovascularization and used deconvolution algorithms to predict the originating cell type of these dysregulated genes. Within the area of neovascularization, endothelial cells demonstrated increased expression of genes related to Rho family GTPase signaling and integrin signaling. Likewise, VEGF and TGFB1 were identified as potential upstream regulators that could drive the observed gene expression changes produced by endothelial and retinal pigment epithelium cells in the macular neovascularization donor. These spatial gene expression profiles were compared to previous single-cell gene expression experiments in human age-related macular degeneration as well as a model of laser-induced neovascularization in mice. As a secondary aim, we investigated regional gene expression patterns within the macular neural retina and between the macular and peripheral choroid. Overall, this study spatially analyzes gene expression across the retina, retinal pigment epithelium, and choroid in health and describes a set of candidate molecules that become dysregulated in macular neovascularization.

Consistency regularisation in varying contexts and feature perturbations for semi-supervised semantic segmentation of histology images

Semantic segmentation of various tissue and nuclei types in histology images is fundamental to many downstream tasks in the area of computational pathology (CPath). In recent years, Deep Learning (DL) methods have been shown to perform well on segmentation tasks but DL methods generally require a large amount of pixel-wise annotated data. Pixel-wise annotation sometimes requires expert’s knowledge and time which is laborious and costly to obtain. In this paper, we present a consistency based semi-supervised learning (SSL) approach that can help mitigate this challenge by exploiting a large amount of unlabelled data for model training thus alleviating the need for a large annotated dataset. However, SSL models might also be susceptible to changing context and features perturbations exhibiting poor generalisation due to the limited training data. We propose an SSL method that learns robust features from both labelled and unlabelled images by enforcing consistency against varying contexts and feature perturbations. The proposed method incorporates context-aware consistency by contrasting pairs of overlapping images in a pixel-wise manner from changing contexts resulting in robust and context invariant features. We show that cross-consistency training makes the encoder features invariant to different perturbations and improves the prediction confidence. Finally, entropy minimisation is employed to further boost the confidence of the final prediction maps from unlabelled data. We conduct an extensive set of experiments on two publicly available large datasets (BCSS and MoNuSeg) and show superior performance compared to the state-of-the-art methods.

SAT571 When Two Is Too Few: Bethesda IV Pathology On Repeat Thyroid Fine Needle Aspiration

Abstract Disclosure: C.M. Godar: None. A.J. Spiro: None. M.K. Shakir: None. T.D. Hoang: None. Background: The 2015 American Thyroid Association Guidelines for Adult Patients with Thyroid Nodules provide a strong recommendation against routine ultrasound surveillance of thyroid nodules for which two benign fine needle aspiration (FNA) cytology results have been previously obtained. Strict adherence to these guidelines may hinder detection of malignancy, especially in large nodules where there is potential to miss heterogeneous pathology based on sampling technique. Clinical Case: A 36-year-old female was evaluated for surveillance of a large non-toxic multinodular goiter. Thyroid ultrasound demonstrated a dominant right thyroid nodule greater than 7 centimeters in size. The patient had undergone FNA of this nodule in 2014 and 2019 with Bethesda II (benign) cytopathology results. A chest x-ray showed tracheal deviation without evidence of narrowing. She noted a sensation of neck pressure without dysphagia or shortness of breath. She denied fatigue, heat or cold intolerance, hand tremor, or pretibial myxedema. She denied a family history of thyroid cancer or any history of neck irradiation. Due to new compressive symptoms a repeat FNA was performed. Three passes were performed with adequate samples, and the final FNA cytopathology demonstrated a Hurthle cell oncocytic-type follicular neoplasm with Bethesda IV classification. The sample was submitted for ThyGeNEXTTM and ThyraMIRv2TM molecular testing, which returned with NRAS q61R mutation of moderate-to-high risk of malignancy (65-75%). The patient was subsequently referred for total thyroidectomy with pending surgery date. Discussion: The current ATA guidelines offer a cost-effective surveillance strategy for the vast majority of patients with thyroid nodules. False negative FNA results are rare and are reported to occur in about 1% of cases1. As in this vignette, it is reasonable to perform an additional FNA after two consecutive Bethesda II results in nodules with clinically significant growth, new compressive symptoms, or if there is concern for incomplete sampling in nodules of large size. During an FNA, care should be taken to gently move the needle at various angles to capture follicular cells from different areas within a nodule. Large nodules should be sampled via multiple, separate skin entry points when feasible. Judicious use of repeat FNA along with these sampling techniques may help to prevent under-diagnosis in nodules containing multifocal areas of heterogeneous pathology. Reference: 1) Durante C, Costante G, Lucisano G, Bruno R, Meringolo D, Paciaroni A, Puxeddu E, Torlontano M, Tumino S, Attard M, Lamartina L, Nicolucci A, Filetti S. 2015. The natural history of benign thyroid nodules. JAMA 313:926-935 Presentation: Saturday, June 17, 2023

Deep learning approach for accurate prostate cancer identification and stratification using combined immunostaining of cytokeratin, p63, and racemase.

Prostate cancer (PCa) is the most frequently diagnosed cancer in men worldwide, affecting around 1.4 million individuals. Current PCa diagnosis relies on histological analysis of prostate biopsy samples, an activity that is both time-consuming and prone to observer bias. Previous studies have demonstrated that immunostaining of cytokeratin, p63, and racemase can significantly improve the sensitivity and the specificity of PCa detection compared to traditional H&E staining. This study introduces a novel approach that combines diagnosis-specific immunohistochemical (IHC) staining and deep learning techniques to provide reliable stratification of prostate glands. Our approach leverages a customized segmentation network, called K-PPM, that incorporates adaptive kernels and multiscale feature integration to enhance the functional information of IHC. To address the high class-imbalance problem in the dataset, we propose a weighted adaptive patch-extraction and specific-class kernel update. Our system achieved noteworthy results, with a mean Dice Score Coefficient of 90.36% and a mean absolute error of 1.64 % in specific-class gland quantification on whole slides. These findings demonstrate the potential of our system as a valuable support tool for pathologists, reducing workload and decreasing diagnostic inter-observer variability. Our study presents innovative approaches that have broad applicability to other digital pathology areas beyond PCa diagnosis. As a fully automated system, this model can serve as a framework for improving the histological and IHC diagnosis of other types of cancer.

Analysis of the Survey Results of Interns of the First Year of Study Regarding Training in the Specialty «Otolaryngology» at the Undergraduate Level

Otolaryngology is a narrow medical specialty that focusses in the diagnosis, treatment and prevention of diseases of the ENT organs, as well as surrounding tissues. Today, there is a need in the experienced doctors who have knowledge, skills and conduct ethical rules. Otolaryngology is an important specialty relevant to family medicine, pediatrics, and general practice, as 20% to 40% of adult general practice consultations are related to otolaryngological complaints. Taking into account the problem of insufficient training in otolaryngology in medical universities in some leading countries, we have proposed a questionnaire for persons who have graduated from medical universities in Ukraine and have not yet completed an internship on the subject of training in the specialty “Otolaryngology” at the university. The objective: to analyze the degree of satisfaction with study of otorhinolaryngology in leading higher educational institutions of Ukraine at the undergraduate level of future general practitioners-family medicine doctors. Materials and methods. A survey of first-year interns who graduated from various medical higher educational institutions of Ukraine was conducted. The questionnaire contained 11 questions. The questionnaire was distributed using Google Forms and processed using Excel. Results. The analysis of the results of the survey showed that the training of doctors in the specialty “Otolaryngology” remains at an insufficient level due to the reduction of teaching hours. 1.9% of respondents were very dissatisfied and 17.3% of the respondents were dissatisfied with the quality of the obtained theoretical knowledge in otorhinolaryngology at the undergraduate level, in fact 1/5 of the respondents. 5.9% and 31.4% of respondents, i.e. 1/3 of surveyed persons, were very dissatisfied and dissatisfied, respectively, with the quality of the acquired practical skills in otorhinolaryngology at the undergraduate level. The majority of respondents (76.9%) plan further training in otorhinolaryngology. Respondents want to improve their knowledge in all areas of ENT pathology (64.7%). Conclusions. The improvement if the training process for future doctors, intensifying the training process and introducing modern methods to it, presenting the latest technology for disease diagnosis and treatment to interns remains a priority. In order to obtain high-quality theoretical and practical skills, it is necessary to increase the time for studying otolaryngology in medical higher education institutions.