A5050449156- Global Cancer Incidence and Screening
- Radiomics and Machine Learning in Medical Imaging
- Hepatocellular Carcinoma Treatment and Prognosis
- Economic and Financial Impacts of Cancer
- Prostate Cancer Diagnosis and Treatment
- Colorectal Cancer Screening and Detection
- AI in cancer detection
- Lymphoma Diagnosis and Treatment
- Cancer Genomics and Diagnostics
- COVID-19 Clinical Research Studies
- Transplantation: Methods and Outcomes
- Viral-associated cancers and disorders
- Cervical Cancer and HPV Research
- Ultrasound and Hyperthermia Applications
- Liver Disease Diagnosis and Treatment
- Breast Cancer Treatment Studies
- Pancreatic and Hepatic Oncology Research
- COVID-19 diagnosis using AI
- Genetic factors in colorectal cancer
- Infection Control and Ventilation
- Molecular Biology Techniques and Applications
- Digital Imaging for Blood Diseases
- Microbial infections and disease research
- Cancer-related molecular mechanisms research
- Acute Lymphoblastic Leukemia research
Heinrich Heine University Düsseldorf
2021-2025
Düsseldorf University Hospital
2021-2025
Martin Luther University Halle-Wittenberg
2018-2024
Luther University
2024
RWTH Aachen University
2021-2023
American Cancer Society
2018
The histopathological phenotype of tumors reflects the underlying genetic makeup. Deep learning can predict alterations from pathology slides, but it is unclear how well these predictions generalize to external datasets. We performed a systematic study on Deep-Learning-based prediction histology, using two large datasets multiple tumor types. show that an analysis pipeline integrates self-supervised feature extraction and attention-based instance achieves robust predictability generalizability.
Abstract Background Monoclonal antibodies (mAbs) that target severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) are predominantly less effective against Omicron variants. Immunocompromised patients often experience prolonged viral shedding, resulting in an increased risk of escape. Methods In observational, prospective cohort, 57 infected with variants who received sotrovimab alone or combination remdesivir were followed. The study end points a decrease SARS-CoV-2 RNA <106...
Abstract Background: Prostate cancer is the leading in men sub-Saharan Africa (SSA) regarding incidence and mortality. Published data from a few registries SSA suggest that rates are still rising, but there little comprehensive information on time trends of prostate incidence. Methods: We analyzed registry 13,170 incident cases aged 40 years or above, 12 population-based 11 countries, with at least 10-year span comparable data. Results: observed an increase cumulative risks (CR)...
One of the most important complications heart transplantation is organ rejection, which diagnosed on endomyocardial biopsies by pathologists. Computer-based systems could assist in diagnostic process and potentially improve reproducibility. Here, we evaluated feasibility using deep learning predicting degree cellular rejection from pathology slides as defined International Society for Heart Lung Transplantation (ISHLT) grading system.
Introduction Assessing the malignancy of focal liver lesions is an important yet challenging aspect routine patient care. Contrast-enhanced ultrasound (CEUS) has proved to be a highly reliable tool but very dependent on examiner's expertise. The emergence artificial intelligence opened doors algorithms that could potentially aid in diagnostic process. In this study, we evaluate performance weakly supervised deep learning model classifying (FLL) as malignant or benign. Methods Our...
Abstract Objectives To estimate observed and relative survival of prostate cancer patients in sub-Saharan Africa (SSA) to examine the influence age, stage at diagnosis Human Development Index (HDI). Patients methods In this comparative registry study, we selected a random sample 1752 incident cases malign prostatic neoplasm from 12 population-based registries 10 SSA countries, registered between 2005 2015. We analyzed data using Kaplan-Meier Ederer II obtain outcome estimates flexible...
Although prostate cancer (PCa) is the most commonly diagnosed in men of sub-Saharan Africa (SSA), little known about its management and survival. The objective current study was to describe presentation, patterns diagnosis, treatment, survival patients with PCa 10 countries SSA.
Abstract Background Cervical cancer (CC) is the most common female in many countries of sub-Saharan Africa (SSA). We assessed treatment guideline adherence and its association with overall survival (OS). Methods Our observational study covered nine population-based registries eight countries: Benin, Ethiopia, Ivory Coast, Kenya, Mali, Mozambique, Uganda, Zimbabwe. Random samples 44–125 patients diagnosed from 2010 to 2016 were selected each. Cancer-directed therapy (CDT) was evaluated for...
Background: Breast cancer (BC) is the most common in sub-Saharan Africa (SSA). However, little known about actual therapy received by women with BC and their survival outcome at population level SSA. This study aims to describe cancer-directed patients SSA, compare these results NCCN Harmonized Guidelines for SSA (NCCN Guidelines), evaluate impact on survival. Methods: Random samples of (≥40 per registry), diagnosed from 2009 through 2015, were drawn 11 urban population–based registries 10...
Non-Hodgkin lymphoma (NHL) is the sixth most common cancer in Sub-Saharan Africa (SSA). Comprehensive diagnostics of NHL are essential for effective treatment. Our objective was to assess frequency subtypes, disease stage and further diagnostic aspects. Eleven population-based registries 10 countries participated our observational study. A random sample 516 patients included. Histological confirmation available 76.2% cytological another 17.3%. subclassification determined 42.1%. Of these,...
Abstract The histopathological phenotype of tumors reflects the underlying genetic makeup. Deep learning can predict alterations from tissue morphology, but it is unclear how well these predictions generalize to external datasets. Here, we present a deep pipeline based on self-supervised feature extraction which achieves robust predictability in two large multicentric datasets seven tumor types.
Abstract The interpretation of digitized histopathology images has been transformed thanks to artificial intelligence (AI). End-to-end AI algorithms can infer high-level features directly from raw image data, extending the capabilities human experts. In particular, predict tumor subtypes, genetic mutations and gene expression hematoxylin eosin (H&E) stained pathology slides. However, existing end-to-end workflows are poorly standardized not easily adaptable new tasks. Here, we introduce...
Abstract We examined trends in childhood cancer incidence sub‐Saharan Africa using data from two population‐based registries Harare (Zimbabwe) and Kyadondo (Uganda) with cases classified according to the International Classification of Childhood Cancer explored reasons for observed variations changes. Over whole 25‐year period (1991‐2015) studied, there were only small, nonsignificant overall incidence. Nevertheless, within period, peaks occurred 1996 2001 2003 2006 (Uganda). Kaposi sarcoma...
The COVID-19 pandemic has been a major burden for healthcare systems worldwide and caused multiple changes problems in outpatient care. aim of this study was to investigate the impact on consultations diagnoses gastroenterology practices Germany. To end, we retrospectively analyzed data from Disease Analyzer database (IQVIA) using International Classification Diseases, 10th revision (ICD-10). We included all patients aged ≥18 years with at least one visit 48 Germany between April September...
Background: The burden of colorectal cancer (CRC) is increasing in Sub-Saharan Africa (SSA). However, little known about CRC treatment and survival the region. Methods: A random sample 653 patients with diagnosed from 2011 to 2015 was obtained 11 population-based registries SSA. Information on clinical characteristics, treatment, and/or vital status medical records treating hospitals for 356 (54%) (“traced cohort”). Concordance NCCN Harmonized Guidelines SSA assessed. Cox proportional...
Abstract Background Although non-Hodgkin lymphoma (NHL) is the 6th most common malignancy in Sub-Saharan Africa (SSA), little known about its management and outcome. Herein, we examined treatment patterns survival among NHL patients. Methods We obtained a random sample of adult patients diagnosed between 2011 2015 from 11 population-based cancer registries 10 SSA countries. Descriptive statistics for lymphoma-directed therapy (LDT) degree concordance with National Comprehensive Cancer...
Purpose Stage at diagnosis and receipt of therapy are the most important determinants breast cancer (BC) survival in sub-Saharan Africa (SSA). Recently, National Comprehensive Cancer Network guidelines for SSA were published. Our study aimed to describe cancer-directed (CDT) received by patients with BC population level SSA. Methods Random samples (≥ 40 cases per registry) who diagnosed from 2009 2015 drawn 11 population-based registries—Abidjan, Addis Ababa, Bamako, Brazzaville, Bulawayo,...
Abstract Background and Aims One of the most important complications heart transplantation is organ rejection, which diagnosed on endomyocardial biopsies by pathologists. Computer-based systems could assist in diagnostic process potentially improve reproducibility. Here, we evaluated feasibility using deep learning predicting degree cellular rejection from pathology slides as defined International Society for Heart Lung Transplantation (ISHLT) grading system. Methods We collected 1079...
Abstract Background To assess population-based quality of cancer care in sub-Saharan Africa and to identify specific gaps joint opportunities, we assessed concordance diagnostics treatments with National Comprehensive Cancer Network Harmonized Guidelines for leading types 10 countries. Methods Adult patients female breast cancer, cervical colorectal non-Hodgkin lymphoma, prostate were randomly drawn from 11 registries. Guideline treatment was using clinical records. In a subcohort 906...
Cervical cancer (CC) has an incidence up to 42/100,000 per year in Sub-Saharan Africa (SSA). Lack of health services, late stages at presentation and poor overall survival (OS) are reported, but little is known about the patterns care. Thus, we designed a retrospective, population-based study collect data upon stage, therapy, OS.