Ovarian cancer is a histologically heterogeneous malignancy in which accurate subtype classification is essential for progno
sis and treatment selection. Manual interpretation of hematoxylin and eosin (H&E)–stained histopathology slides remain
time-consuming and subject to inter-observer variability, particularly for morphologically overlapping subtypes. In this
study, an interpretable deep learning framework was developed for automated ovarian cancer subtype classification using a
fine-tuned ResNet50 architecture. A publicly available histopathology dataset comprising five major ovarian carcinoma sub
types was employed. Model training incorporated optimized strategies including data augmentation, selective layer unfreez
ing, label smoothing, and test-time augmentation. Classification performance was benchmarked against established convolu
tional neural network architectures. Visual interpretability was assessed using Gradient-weighted Class Activation Mapping
(Grad-CAM) to examine model attention patterns.
Indexing
Top Links
E-mail: mail@annexpublishers.co
Radical Novelty of Research
NIH Funded Articles 
- Blistering Disease in an End Stage Renal Disease Patient
- Evaluation of Volume Reduction Using the Kuhnke Method after Instrumental Lymphatic Drainage in Patients with Stage I and II Lymphedema: A Three-Month Study in Moscow
- Predictors of Obstructive Sleep Apnea Risk among Blacks with Metabolic Syndrome
- Molecular Mechanism Linking BRCA1 Dysfunction to High Grade Serous Epithelial Ovarian Cancers with Peritoneal Permeability and Ascites
- Reasons for hospitalization of HIV-positive patients in Riga East clinical University hospital, Latvia, 2014
- Penicillin-Streptomycin Induces Mouse Embryonic Stemcells Differentiation into Cardiomyocytes Through MAPK Signal Pathway
- Development of Acute Myeloid Leukemia Following CDK4/6 Inhibitor Therapy for Patient with Metastatic Hormone Receptor Positive Breast Cancer (mHR+BC) and Pre-existing Clonal Hematopoiesis of Indeterminate Potential–A Case Report
- Estimation of Glomerular Filtration Rate Comparison of Equations to Estimate GFR: Cockcroft-Gault, MDRD, CKD - Epi and BIS1
- Ricinine Toxicity Presenting as Acute Flaccid Paralysis
- Wake-up Strokes Are Similar to Known-Onset Morning Strokes in Severity and Outcome
-
Annex Publishers
Annex publishers is an Open Access publication model which allows the publication of articles to the world through this community. We provide you the asset of interaction with the most researchers from the scientific community. Anyone can freely access the articles printed under Annex publishers. -
Why Open Access?
Open access provides an unlimited access to journals and publications, with a perfect platform for thousands of researchers to form jointly for the development of science. Open Access assists in accelerating the scientific discovery, encourage innovation, enhance education and refresh the economy to build up the knowledge of the general public.
Submit Manuscript
Search here
Recent Articles
Media Influence on Childhood Obesity in Lahore, Pakistan Affiliation
Media strongly influences children’s eating habits, promoting consumption of sugary and ultra-processed foods.
Diagnosis of the Incidence of Fusarium Wilt on Tomato in Jaba Local Goverment Area. Kaduna State
Fusarium wilt disease of tomato is caused by Fusarium oxysporum f.sp. lycopersici and is a limiting factor to tomato production in Nigeria.
HIV Prevalence Trends and Risk Behaviors among Married Women, Mozambique: Analysis of the 2015 and 2021 National AIDS Indicator Surveys (IMASIDA and INSIDA)
Sub-Saharan Africa has experienced substantial declines in new HIV infections over the past decade; however, these gains have not been uniform across populations. In Mozambique, married women remain a largely overlooked group in HIV prevention strategies, despite persistently high HIV prevalence.
