Morphological Classification Research Articles

LW-MorphCNN: a lightweight morphological attention-based subtype classification network for lung cancer

Abstract Lung cancer is generally considered one of the most deadly cancers globally. If it can be identified early and diagnosed correctly, the survival probability of patients can be significantly improved. In this process, histopathological examination is a commonly used method for diagnosing and detecting lung cancer. It is crucial to accurately identify lung cancer subtypes from histopathological images, as this helps doctors formulate effective treatment plans. However, the visual inspection in histopathological diagnosis requires a large amount of time and also depends on the subjective perception of clinicians. Therefore, this paper proposes a lightweight lung cancer subtype classification network based on morphological attention (LW-MorphCNN), which is used to automatically classify the histopathological images of benign lung tumors, ADC (adenocarcinoma), and SCC (squamous cell carcinoma) provided in the public dataset LC25000 (Lung and Colon). This paper takes histopathological images as input and conducts a comparative analysis with classic networks such as VGG16, VGG19, DenseNet121, and ResNet50, as well as existing classification methods proposed in the same work. The network proposed in this paper is superior to other networks in terms of parameter quantity and performance, with an accuracy rate and F1 - score reaching 99.47% and 99.44% respectively. Clinicians can install the provided LW-MorphCNN in the hospital to confirm the diagnosis results.

Tricuspid regurgitation: a contemporary review.

Tricuspid regurgitation (TR), an underrecognized disease, overlooked by clinicians for many years due to its assumed benign nature. Recent epidemiological studies suggest significant TR may be seen in up to 6% of elderly patients. An increase in prevalence is expected due to the higher incidence of various clinical predictors of TR progression. Increasing severity of TR is associated with worse outcomes with a novel morphologic classification providing a more refined prediction of outcomes. Advances in cardiac imaging, particularly echocardiography, are integral to the diagnosis of disease severity which not only includes quantitation of TR, but also an assessment of the right atrium, right ventricle and pulmonary arterial circulation. Once identified and quantified, TR management requires a multi-disciplinary heart team management including structural imagers, heart failure specialists, electrophysiologist, cardiac surgeons and interventionalists. Data to support medical therapies are lacking although guidelines support the management of congestive signs and symptoms, as well as comorbidities such as left heart failure and rhythm management. The risks of surgical interventions are slowly improving, however, transcatheter therapies are now available to treat patients with high surgical risk. This manuscript will provide a state-of-art review of this fast-moving field, including current scientific evidences, but also upcoming perspectives with multiple ongoing clinical studies.

Novel morphological classification of left atrial appendage using 3-dimensional printer -rock, paper and scissors-

Abstract Background The WATCHMAN device was developed as a permanent implantable device to seal off the LAA to prevent cardiogenic embolism. It has been used in Japan since September 2019 only for patients who are at the high risk of bleeding events. It is known that the LAA diameter of Asians is larger than that of non-Asians. Furthermore, the LAA of the Japanese are more complicated, compared to the Western. Whereas the four classifications from previous reports are widely used in the world, the classification differs with different physicians in the real-world practice. Purpose We aimed to identify the relationship between morphological features of LAA and clinical events such as ischemic stroke in Japanese patients. Methods All LAAC cases in our institution performed between February 2020 and December 2023 were retrospectively analyzed. The three-dimensional printing of LAA was generated from cardiovascular computed tomography imaging data. The shapes of the LAA were classified into 3 morphological types: (1) The LAA with bi lobe was defined as "Scissors" type; (2) The one with complicated peripheral branches was "Paper" type; (3) The simple one was "Rock" type. We examined the between the risk of cardioembolic stroke and the morphological features of the LAA. The left atrial appendage was classified by two cardiologists, and when they made different choices, we used the operator's choice. Results The cohort consisted of 116 patients, excluding 22 patients without CT scans. Of those, 30 (26%), 38 (33%) and 48 (41%) patients were divided into the Rock, Scissors and Paper types, respectively. All groups were identically matched for comorbidities. Table shows the patient characteristics and the results of univariate and multivariate analysis. There were 51 (44%) patients who had a history of cardioembolic stroke (18% vs. 30% vs. 53%; Rock vs. Scissors vs. Paper types, p-value = 0.05). In the multivariate analysis, the Paper type is the potential risk factor of cardioembolic stroke (p-value = 0.01). In addition, of those, 36 (31%), 32 (28%), 28 (24%) and 20 (17%) patients were divided into the Chicken wing, Cactus, Cauliflower and Windsock types, respectively. The four types were not associated with prior cardioembolic stroke events (18% vs. 25% vs. 55%; Chicken wing vs. Cactus vs. Cauliflower vs. Windsock types, p-value = 0.83). The match rate of cardiologists’ choice was 86.2% for the new classification and 44.8% for the classical classification. Conclusions The morphological features of LAA by three-dimensional printing was significantly related to cardioembolic stroke in this population. Patients with Paper type LAA were significantly more likely to have had a history of cardioembolic stroke. These results could have a relevant impact on the antithrombotic management of patients who are at the high risk of embolic and bleeding events and, moreover, LAAC could be the best option for these patients.

Effects of urban green space habitats and tree species on ectomycorrhizal fungal diversity

Ectomycorrhizal fungi (EMF) are key symbiotic microbial components for the growth and health of trees in urban greenspace habitats (UGSHs). However, the current understanding of EMF diversity in UGSHs remains poor. Therefore, in this study, using morphological classification and molecular identification, we aimed to investigate EMF diversity in three EMF host plants: Cedrus deodara in the roadside green belt, and C. deodara, Pinus massoniana, and Salix babylonica in the park roadside green belt, in Guiyang, China. A total of 62 EMF Operational Taxonomic Units (OTUs) were identified, including 13 EMF OTUs in the C. deodara roadside green belt, and 23, 31, and 9 EMF OTUs in the park green belts. C. deodara, P. massoniana, and S. babylonica were respectively identified in park green belts. Ascomycota and Basidiomycota were the dominant phylum in the EMF communities in roadside and park green habitat, respectively. The Shannon and Simpson indexes of the C. deodara EMF community in the park green belt were higher than those in the roadside green belt. EMF diversity of the tree species in the park green belt was P. massoniana > C. deodara > S. babylonica. Differences in EMF community diversity was observed among the different greening tree species in the UGSHs. UGSHs with different disturbance gradients had a significant impact on the EMF diversity of the same greening tree species. These results can be used as a scientific reference for optimizing the design and scientific management of UGSHs.

Optimization of Steaming Conditions for Bellflower Root (Platycodon grandiflorus) Using K-Means Clustering-Based Morphological Grading System

Bellflower roots were categorized into three clusters (class 0, class 1, and class 2) using K-means clustering based on their morphological factors: length (282.8 ± 29.53, 138.75 ± 26.8, and 209.89 ± 20.49 mm), thickness (16.25 ± 2.82, 16.77 ± 3.35, and 16.52 ± 3.05 mm), and body shape coefficient (5.80 ± 1.15, 12.73 ± 4.82, and 7.95 ± 1.71). Internal void formation, a key quality factor for bellflower root, was analyzed under pre-steaming conditions, identifying temperatures between 20 and 25 °C as optimal for storage. Within the clustered class, steaming for a prolonged duration increased the formation of internal voids and caused a decrease in normal stress values, total dissolved solids (TDS), and pectin content. Class 0, with larger and thicker roots, exhibited higher internal voids (57% void rate) due to uneven heat distribution and incomplete starch gelatinization. Class 2 roots demonstrated better structural integrity, with a void rate of 26% and a stress value of 48 kN/m2. These findings highlight the importance of morphological classification and optimal storage temperatures to improve the quality of steamed bellflower roots.

Stent Underexpansion Is an Underestimated Cause of Intrastent Restenosis: Insights From RESTO Registry.

Despite improvement in devices, in-stent restenosis remains a frequent and challenging complication of percutaneous coronary interventions. The RESTO (Morphological Parameters of In-Stent Restenosis Assessed and Identified by OCT [Optical Coherence Tomography]; study NCT04268875) was a prospective multicenter registry including patients presenting with coronary syndromes related to in-stent restenosis. All patients underwent preintervention OCT analysis, which led to analysis of in-stent restenosis phenotype, number of strut layers, and presence of stent underexpansion. The primary end point was the in-stent restenosis type according to the OCT morphological classification. The 1-year incidence of target vessel failure (a composite of death from cardiac causes, target-vessel myocardial infarction, or ischemia-driven target-vessel revascularization) was assessed. The study included 297 patients. The culprit stent was a drug-eluting stent in 74.2% of cases. OCT analysis revealed the presence of neoatherosclerosis in 57% (52% calcified), neointimal hyperplasia in 43% (58% homogeneous), stent underexpansion (minimal stent area <4.5 mm2) in 43%, and multiple stent layers in 30%. The prepercutaneous coronary intervention OCT analysis modified the operator's strategy for management in 30% of cases. Treatment involved drug-eluting stent implantation in 61.6% and drug-eluting balloon angioplasty in 36.1% of cases with only 63.2% optimal results. The 1-year target vessel failure incidence was 11% (95% CI, 9%-13%). Residual postpercutaneous coronary intervention stent underexpansion was associated with significantly higher target vessel failure incidence (19% [95% CI, 14%-24%] versus 7% [95% CI, 5-9], P=0.01). OCT identified neoatherosclerosis and neointimal hyperplasia in comparable proportions. Stent underexpansion was frequent and favored subsequent adverse clinical outcomes.

Experimental Study on the Classification and Evolution of the Tip Cavitation Morphology in Axial Waterjet Pumps with Two Different Blade Numbers

Tip leakage flow and induced unstable cavitation can significantly damage the performance of axial waterjet pumps. This study investigated the impact of blade numbers on cavitating conditions in an axial waterjet pump by conducting tests of performance characteristics and high-speed photography experiments on three-blade and four-blade impellers. The results showed that the critical cavitation number σc of the three-blade impeller was larger, while the four-blade impeller flow pattern deteriorated more rapidly after σc. Various cavitation structures in the tip region were observed under different conditions, including clearance cavitation, shear layer cavitation, tip leakage vortex cavitation, and suction-side-perpendicular cavitating vortices (SSPCVs). Tip cavitation maps of the test impellers were drawn based on the flow rate coefficient and cavitation number variation. The three-blade impeller exhibited a wider range of severe cavitation, particularly with an increased occurrence of SSPCVs. With the cavitation number and flow rate coefficient decreased, the SSPCV generated from triangular cavitation cloud shedding presented an increased trend in scale and quantity. Conversely, in the case of the four-blade impeller, SSPCVs were often disrupted by the adjacent blade during migration and interfered with the tip cavitation in the neighboring flow passage.

Validation of morphological ear classification devised by principal component analysis using three-dimensional images for human identification.

This study attempts to classify ear morphology for human identification in forensic investigations by distinguishing between the upper auricle and lobule areas. A three-dimensional homologous model of the ear was created using 414 ear images of males aged 17-93 years reconstructed from computed tomography scans of forensic autopsy cases. Morphological changes were visualized using principal component analysis and areas of significant individual differences within the entire ear were identified. The classification criterion images for the upper auricle (ten images) and lobule (12 images) were developed by combining multiple principal component values: components 1-5 for the upper auricle and 1-6 for the lobule. Three-dimensional ear images of the upper auricle and lobule areas from 414 subjects were categorized using a measurement method based on the minimum distance between 5,507 corresponding points. The results indicate the applicability of the criterion images for the morphological classification of ears in this study.

Success Rate of Drilling and Knuckle Wire Crossing Technique in Chronic total Occlusion of the Femoro-Popliteal Artery and Predicting Factors Related to it in Real-World Data

Objective: Crossing chronic total occlusion (CTO) in femoro-popliteal (FP) arteries is challenging. Hence, this study aimed to describe real-world data on the technical success rate of crossing FP CTO lesions, and analyze predicting factors for crossing failure. Material and Methods: This retrospective study, conducted at a single center, involved patients with peripheral arterial disease (PAD) and FP CTOs, categorized as Rutherford categories 3-6, having undergone endovascular recanalization from 2017 to 2022. Baseline characteristics and crossing success rates were described using frequencies, percentages, means, medians, standard deviations, and interquartile ranges. Factors associated with crossing failure were analyzed through logistic regression analysis, with statistical significance set at a p-value&lt;0.05.Results: Among the 181 patients analyzed, the majority exhibited minor tissue loss (63%) and severe disease (TASC II category D, GLASS grade 4). The predominant CTO crossing approach was based on plaque cap morphology (CTOP) classification type B (66.9%), with severe CTO calcification (PACSS type 4) in 23.8%. Median occlusion length was 16.1 cm; 85.6% successfully crossed the CTO and antegrade success was 57.5%. Factors associated with crossing failure included: CTOP types C or D, CTO at the P3-popliteal artery, severe CTO calcification, and dyslipidemia. Retrograde recanalization showed higher success rates in CTOP type D and longer occlusions.Conclusion: This study highlights the safety and effectiveness of the catheter-wire crossing technique for FP CTOs. CTOP classification, CTO location, severe calcification, and dyslipidemia predicted crossing failure. The retrograde approach works well in CTOP type D and longer CTOs, providing insights for improving success in clinical practice.

At First Sight! Zero-shot Classification of Astronomical Images with Large Multimodal Models

Abstract Vision-Language multimodal Models (VLMs) offer the possibility for zero-shot classification in astronomy: i.e., classification via natural language prompts, with no training. We investigate two models, GPT-4o and LLaVA-NeXT, for zero-shot classification of low-surface brightness galaxies and artifacts, as well as morphological classification of galaxies. We show that with natural language prompts these models achieved significant accuracy (above 80% typically) without additional training/fine tuning. We discuss areas that require improvement, especially for LLaVA-NeXT, which is an open source model. Our findings aim to motivate the astronomical community to consider VLMs as a powerful tool for both research and pedagogy, with the prospect that future custom-built or fine-tuned models could perform better.

Multi-level feature fusion network for neuronal morphology classification.

Neuronal morphology can be represented using various feature representations, such as hand-crafted morphometrics and deep features. These features are complementary to each other, contributing to improving performance. However, existing classification methods only utilize a single feature representation or simply concatenate different features without fully considering their complementarity. Therefore, their performance is limited and can be further improved. In this paper, we propose a multi-level feature fusion network that fully utilizes diverse feature representations and their complementarity to effectively describe neuronal morphology and improve performance. Specifically, we devise a Multi-Level Fusion Module (MLFM) and incorporate it into each feature extraction block. It can facilitate the interaction between different features and achieve effective feature fusion at multiple levels. The MLFM comprises a channel attention-based Feature Enhancement Module (FEM) and a cross-attention-based Feature Interaction Module (FIM). The FEM is used to enhance robust morphological feature presentations, while the FIM mines and propagates complementary information across different feature presentations. In this way, our feature fusion network ultimately yields a more distinctive neuronal morphology descriptor that can effectively characterize neurons than any singular morphological representation. Experimental results show that our method effectively depicts neuronal morphology and correctly classifies 10-type neurons on the NeuronMorpho-10 dataset with an accuracy of 95.18%, outperforming other approaches. Moreover, our method performs well on the NeuronMorpho-12 and NeuronMorpho-17 datasets and possesses good generalization.

Multi-Task Learning Network-Based Prediction of Hydraulic Fracturing Effects in Horizontal Wells Within the Ordos Yanchang Formation Tight Reservoir

Accurate prediction of fracture volume and morphology in horizontal wells is essential for optimizing reservoir development. Traditional methods struggle to capture the intricate relationships between fracturing effects, geological variables, and operational factors, leading to reduced prediction accuracy. To address these limitations, this paper introduces a multi-task prediction model designed to forecast fracturing outcomes. The model is based on a comprehensive dataset derived from fracturing simulations within the Long 4 + 5 and Long 6 reservoirs, incorporating both operational and geological factors. Pearson correlation analysis was conducted to assess the relationships between these factors, ranking them according to their influence on fracturing performance. The results reveal that operational variables predominantly affect Stimulated Reservoir Volume (SRV), while geological variables exert a stronger influence on fracture morphology. Key operational parameters impacting fracturing performance include fracturing fluid volume, total fluid volume, pre-fluid volume, construction displacement, fracturing fluid viscosity, and sand ratio. Geological factors affecting fracture morphology include vertical stress, minimum horizontal principal stress, maximum horizontal principal stress, and layer thickness. A multi-task prediction model was developed using random forest (RF) and particle swarm optimization (PSO) methodologies. The model independently predicts SRV and fracture morphology, achieving an R2 value of 0.981 for fracture volume predictions, with an average error reduced to 1.644%. Additionally, the model’s fracture morphology classification accuracy reaches 93.36%, outperforming alternative models and demonstrating strong predictive capabilities. This model offers a valuable tool for improving the precision of fracturing effect predictions, making it a critical asset for reservoir development optimization.

Identification of New Cultivar and Different Provenances of Dendrocalamus brandisii (Poaceae: Bambusoideae) Using Simple Sequence Repeats Developed from the Whole Genome.

Dendrocalamus brandisii is a high-quality bamboo species that can be used for both bamboo shoots and wood. The nutritional components and flavors of D. brandisii vary from different geographical provenances. However, the unique biological characteristics of bamboo make morphological classification methods unsuitable for distinguishing them. Although the new cultivar 'Manxie No.1' has significant differences in the branch characteristics and the color of shoot sheaths compared to the D. brandisii, it still lacks precise genetic information at the molecular level. This study identified 231,789 microsatellite markers based on the whole genome of D. brandisii and analyzed their type composition and distribution on chromosomes in detail. Then, using TP-M13-SSR fluorescence-labeling technology, 34 pairs of polymorphic primers were screened to identify the new cultivar 'Manxie No.1' and 11 different geographical provenances of D. brandisii. We also constructed DNA fingerprinting profiles for them. At the same time, we mapped six polymorphic SSRs to the gene of D. brandisii, among which SSR673 was mapped to DhB10G011540, which is related to plant immunity. The specific markers selected in this study can rapidly identify the provenances and the new cultivar of D. brandisii and help explore candidate genes related to some important traits.

Assessment of Trichoderma Species Isolated From Volcanic Soil of a Durian Field in Sisaket Province, Thailand for Plant Growth Promotion and Biocontrol Potential

Trichoderma species are ubiquitous saprophytic fungi commonly found in soil, with potential as fungal biocontrol agents for plant disease control and growth promotion. This research aimed to assess the attributes of Trichoderma sp. as plant growth promoters, focusing on nitrogen fixation, siderophore production, plant nutrient solubilization and indole-3-acetic acid production. Out of 19 isolates tested, 6 (T05, T19, T25, T27, T28 and T33) showed superior plant growth promotion abilities. These isolates were further evaluated for their capacity to enhance the germination of Khao Dawk Mali 105 rice seeds. Seeds were soaked in a suspension of conidia at a concentration of 1×1012 conidia/mL, and germination rates were measured. The germination rate ranged between 77.67 - 86.67 %, statistically significantly higher than the control rate of 33.33 %. Among the isolates, T05 exhibited the highest promotion of plant growth, evidenced by root length (7.27 cm), shoot length (8.50 cm), fresh weight (0.54 g) and dry weight (0.27 g), all significantly different from the control. Additionally, the potential of isolates to inhibit the fungal pathogen Colletotrichum sp. MN-3 was assessed using a dual culture method, with T35 showing the highest inhibition percentage (60.90 %). Morphological classification and nucleotide sequencing of the ITS1-5.8S-ITS2 region of rDNA gene identified the isolates as T. harzianum, T. reesei, T. asperellum and T. longibrachiatum. These findings underscore the effectiveness of Trichoderma sp. in promoting plant growth and suppressing plant pathogenic fungi. HIGHLIGHTSSix Trichoderma isolates (T05, T19, T25, T27, T28 and T33) exhibited superior plant growth promotion abilities. These isolates significantly enhanced germination rates of Khao Dawk Mali 105 rice seeds compared to the method. Among the isolates, T05 showed the greatest promotion of plant growth, as evidenced by root length, shoot length, fresh weight and dry weight. Trichoderma isolates demonstrated potential for inhibiting the fungal pathogen Colletotrichum sp. MN-3. Morphological classification divided the Trichoderma sp. into 4 groups, and species identification through nucleotide sequencing revealed T. harzianum, T. reesei, T. asperellum and T. longibrachiatum. GRAPHICAL ABSTRACT

Application of multimodal intraoperative neurophysiological monitoring technology in neurofibromatosis type 1 related peripheral nerve tumor surgery

To summarize application effect and clinical experience of multimodal intraoperative neurophysiological monitoring (IONM) technology in the surgery of neurofibromatosis type 1 (NF1) related peripheral nerve tumors. A retrospective study was conducted on NF1 patients, who admitted between January 2019 and December 2023 and treated with peripheral nerve tumor resection surgery assisted by multimodal IONM technology. There were 49 males and 45 females. The age ranged from 5 to 78 years, with an average of 33.7 years. Tumor morphological classification included 71 cases of nodular type, 13 cases of diffuse type, and 10 cases of mixed type. Target tumors were distributed in craniofacial region (47 cases), neck (11 cases), trunk (12 cases), and limbs (24 cases). Preoperatively, 44 cases had no obvious neurological symptoms, while the remaining patients had neurological symptoms, including 15 cases of visual impairment, 5 cases of hearing impairment, 16 cases of somatic movement disorders, and 31 cases of somatic sensory disorders, of which 7 cases had more than one symptom. IONM plans were selected based on the relevant nerves and adjacent important structures of the target tumor, including visual evoked potential (17 cases), somatosensory evoked potential (44 cases), motor evoked potential (88 cases), and electromyogram (94 cases). All surgeries were successfully completed. Ninety-three patients underwent total/near total resection and 1 patient underwent palliative resection. Pathological examination showed 80 cases of neurofibroma and 14 cases of malignant peripheral nerve sheath tumors. Complications included 2 cases of hematoma and 3 cases of incision infection. All patients were followed up 3-61 months (median, 15 months). During follow-up, no significant changes in neurological symptoms or tumor recurrence were found. Among the patients with preoperative visual impairment, there were 14 cases with no improvement in symptoms and 1 with improvement after surgery. Among the patients with somatic movement disorders, there were 11 cases with no improvement in symptoms, 3 cases with improvement, 2 cases with aggravation, 4 newly onset cases, and 1 case with significant impact on daily life after surgery. Among the patients with somatic sensory disorders, there were 17 cases with no improvement in symptoms, 14 cases with improvement, and 13 newly onset cases. The patients with hearing impairment showed improvement after surgery. The clinical manifestations of NF1 related peripheral nerve tumors are complex. Multimodal IONM technology can provide real-time detection of nerve provocation and damage. Surgical treatment with multimodal IONM technology is safe and can reduce complications.

Galaxy Classification Using EWGC

The Enhanced Wide-field Galaxy Classification Network (EWGC) is a novel architecture designed to classify spiral and elliptical galaxies using Wide-field Infrared Survey Explorer (WISE) images. The EWGC achieves an impressive classification accuracy of 90.02%, significantly outperforming the previously developed WGC network and underscoring its superior performance in galaxy morphology classification. Remarkably, the network demonstrates a consistent accuracy of 90.02% when processing both multi-target and single-target images. Such robustness indicates the EWGC’s versatility and potential for various applications in galaxy classification tasks.

Real-time monitoring of urban roadway health: Utilizing GPR techniques for the early detection and classification of subsurface cavity diseases

The effectiveness of ground penetrating radar (GPR) in identifying and categorizing diseases that occur underground beneath the surfaces of urban roads is investigated in this study. Both 2D and 3D forward modeling use simulation with the GprMax program to show the response characteristics of common cavity illnesses, which facilitates interpretation in practical situations. The cavity morphology classification accuracy is improved to 90.5% by using convolutional neural networks (CNNs), specifically transfer learning with AlexNet. This method outperforms existing approaches even with minimal data. Four primary types are identified from an analysis of 1965 subsurface cavity data: hollow bodies, empty bodies, loose bodies, and water-rich bodies. These categories are important for evaluating road risks such as voids and subsidence. However, it is still difficult to interpret picture features linked to cavity diseases accurately because of a variety of elements, such as anthropogenic, environmental, and geological influences. However, the accurate interpretation and recognition of image features related to cavity diseases remain challenging. Moreover, there are various factors involved in the formation of underground diseases and cavities, including geological and environmental factors, physical and chemical properties of the geotechnical materials, anthropogenic engineering activity and social population or commercial effects.

Urban Morphology Classification and Organizational Patterns: A Multidimensional Numerical Analysis of Heping District, Shenyang City

Prior studies have failed to adequately address intangible characteristics and lacked a comprehensive quantification of cultural dimensions. Additionally, such works have not merged supervised and unsupervised classification methodologies. To address these gaps, this study employed multidimensional numerical techniques for precise spatial pattern recognition and urban morphology classification at the block scale. By examining building density, mean floor numbers, functional compositions, and street block mixed-use intensities, alongside historical and contemporary cultural assets within blocks—with assigned weights and entropy calculations from road networks, building vectors, and POI data—a hierarchical categorization of high, medium, and low groups was established. As a consequence, cluster analysis revealed seven distinctive morphology classifications within the studied area, each with unique spatial configurations and evolutionary tendencies. Key findings include the dominance of high-density, mixed-use blocks in the urban core, the persistence of historical morphologies in certain areas, and the emergence of new, high-rise clusters in recently developed zones. The investigation further elucidated the spatial configurations and evolutionary tendencies of each morphology category. These insights lay the groundwork for forthcoming studies to devise morphology-specific management strategies, thereby advancing towards a more scientifically grounded, rational, and precision-focused approach to urban morphology governance.

A-156 Comparison of blast and variant lymphocyte flagging between the Alinity hq from Abbott and XN-1000 from Sysmex

Abstract Background Blast cells are immature white blood cells (WBC). The presence of blasts in the peripheral blood may be associated with certain diseases such as myelodysplastic syndrome and acute myelogenous leukemia. Variant lymphocytes (i.e., reactive lymphocytes, atypical lymphocytes) are WBCs produced by the immune system in reaction to infections caused by pathogens such as Epstein-Barr virus, cytomegalovirus, etc. It is clinically useful that the automated hematology analyzer is capable of flagging these cells. In this study, the blast (shown as ‘BLAST’ on the Alinity hq and ‘Blast/Abn Lympho?’ or ‘Blast?’ on XN-1000) and variant lymphocyte (shown as ‘VAR LYM’ on the Alinity hq, and ‘Atypical Lympho?’ on XN-1000) flagging was compared between the Alinity hq from Abbott and XN-1000 from Sysmex. Alinity hq has different sensitivity settings available for BLAST and VAR LYM flagging. In this study, the default setting for Alinity hq was used. For XN-1000, the Q-Flag default setting is 100. In this study, the setting was 120 for ‘Blast/Abn Lympho’ and 100 for “Atypical lympho?’. Methods Blood specimens were tested on the Alinity hq and XN-1000 analyzers. Two smears per specimen were made manually or by the Alinity hs slide maker stainer and read by two independent readers. The specimens showing any of the following flagging were included in the analysis: ‘BLAST’ on Alinity hq, ‘VAR LYM’ on Alinity hq, ‘Blast/Abn Lympho?’ or ‘Blast?’ on XN-1000, ‘Atypical lympho?’ on XN-1000. The flagging reports of these samples from both analyzers were compared to the smear review to determine if each analyzer's blast and variant lymphocyte flagging were true positive, false positive, true negative, or false negative. Results A total of 520 specimens were tested, yielding 41 specimens for blast flagging analysis and 80 specimens for variant lymphocyte flagging analysis. Due to the scarce availability of true positive samples from the manual review, true positive and false negative rate were not compared. For blast flagging using smear results of &amp;gt;=1% or &amp;gt;=5% as the positive criterion, Alinity hq yielded a 9.8% (=4/41) false positive rate and 87.8% (=36/41) true negative rate. XN-1000 yielded a 75.6% (=31/41) false positive rate and 22.0% (=9/41) true negative rate. For variant lymphocyte flagging using smear results of &amp;gt;=5% as the positive criterion, Alinity hq yielded a 6.3% (=5/80) false positive rate and 92.5% (=74/80) true negative rate. XN-1000 yielded a 91.3% (=73/80) false positive rate and 7.5% (=6/80) true negative rate. Conclusions This study demonstrated that Alinity hq achieved a much lower false positive rate and much higher true negative rate for both blast and variant lymphocyte flagging when compared to the XN-1000. Manual blast and variant lymphocyte morphology classification is laboratory- and technician-dependent. Both analyzers allow different sensitivity/specificity levels to flag blasts and variant lymphocytes. Each individual laboratory should evaluate and choose the setting that is most suitable for their specific patient populations to obtain the desired flagging rate.

Self-supervised learning on MeerKAT wide-field continuum images

Context. Self-supervised learning (SSL) applied to natural images has demonstrated a remarkable ability to learn meaningful, low-dimension representations without labels, resulting in models that are adaptable to many different tasks. Until now, applications of SSL to astronomical images have been limited to Galaxy Zoo datasets, which require a significant amount of preprocessing to prepare sparse images centered on a single galaxy. With wide-field survey instruments at the forefront of the Square Kilometer Array (SKA) era, this approach to gathering training data is impractical. Aims. We demonstrate that continuum images from surveys such as the MeerKAT Galactic Cluster Legacy Survey (MGCLS) can be successfully used with SSL, without extracting single-galaxy cutouts. Methods. Using the SSL framework DINO, we experimented with various preprocessing steps, augmentations, and architectures to determine the optimal approach for this data. We trained both ResNet50 and Vision Transformer (ViT) backbones. Results. Our models match state-of-the-art results (trained on Radio Galaxy Zoo) for FRI/FRII morphology classification. Furthermore, they predict the number of compact sources via linear regression with much higher accuracy. Open-source foundation models trained on natural images such as DINOv2 also excel at simple FRI/FRII classification; the advantage of domain-specific backbones is much smaller models trained on far less data. Smaller models are more efficient to fine-tune, and doing so results in a similar performance between our models, the state-of-the-art, and open-source models on multi-class morphology classification. Conclusions. Using source-rich crops from wide-field images to train multi-purpose models is an easily scalable approach that significantly reduces data preparation time. For the tasks evaluated in this work, twenty thousand crops is sufficient training data for models that produce results similar to state-of-the-art. In the future, complex tasks like source detection and characterization, together with domain-specific tasks, ought to demonstrate the true advantages of training models with radio astronomy data over natural-image foundation models.