A5044173766- Cervical Cancer and HPV Research
- AI in cancer detection
- Genital Health and Disease
- Global Cancer Incidence and Screening
- Molecular Biology Techniques and Applications
- Radiomics and Machine Learning in Medical Imaging
- Hepatitis B Virus Studies
- Colorectal Cancer Screening and Detection
- Reproductive tract infections research
- Endometrial and Cervical Cancer Treatments
- Vaccine Coverage and Hesitancy
- Medical Imaging and Analysis
- Cancer Risks and Factors
- Machine Learning in Healthcare
- Viral-associated cancers and disorders
- Artificial Intelligence in Healthcare and Education
- Medical Imaging Techniques and Applications
- Biomedical Text Mining and Ontologies
- Artificial Intelligence in Healthcare
Information Management Services
2015-2025
University of Washington
2021-2025
Seattle University
2024
National Cancer Institute
2010-2023
Division of Cancer Epidemiology and Genetics
2018-2023
Engineering Service Center und Handel (Germany)
2019
Anna Needs Neuroblastoma Answers
2019
Kaiser Permanente
2017-2018
Bayer (United States)
2018
Becton Dickinson (United States)
2018
Human papillomavirus vaccination and cervical screening are lacking in most lower resource settings, where approximately 80% of more than 500 000 cancer cases occur annually. Visual inspection the cervix following acetic acid application is practical but not reproducible or accurate. The objective this study was to develop a "deep learning"-based visual evaluation algorithm that automatically recognizes precancer/cancer.A population-based longitudinal cohort 9406 women ages 18-94 years...
The 2019 American Society for Colposcopy and Cervical Pathology Risk-Based Management Consensus Guidelines the management of cervical cancer screening abnormalities recommend 1 6 clinical actions (treatment, optional treatment or colposcopy/biopsy, 1-year surveillance, 3-year 5-year return to regular screening) based on risk intraepithelial neoplasia grade 3, adenocarcinoma in situ, (CIN 3+) many different combinations current recent past results. This article supports main guidelines...
HPV testing is replacing cytology for cervical cancer screening because of greater sensitivity and superior reassurance following negative tests the dozen genotypes that cause cancer. Management women positive unresolved. The need identification individual clinical use debated. Also, it unclear how long to observe persistent infections when precancer not initially found.
Human papillomavirus (HPV) types differ profoundly in cervical carcinogenicity. For the most carcinogenic type HPV16, variant lineages representing further evolutionary divergence also cancer risk. Variants of remaining 10 to 15 HPV have not been well studied. In first prospective, population-based study variants, we explored whether, on average, oldest branches within each predicted different risks >2-year viral persistence and/or precancer and [cervical intraepithelial neoplasia grade 3+...
The main goal of cervical screening programs is to detect and treat precancer before cancer develops. Human papillomavirus (HPV) testing more sensitive than cytology for detecting precancer. However, reports rare HPV-negative, cytology-positive cancers are motivating continued use both tests (cotesting) despite increased costs. We quantified the detection by cotesting compared with HPV alone at Kaiser Permanente Northern California (KPNC), where 1 208 710 women age 30 years older have...
Purpose Obesity has been inconsistently linked to increased cervical cancer incidence and mortality; however, the effect of obesity on screening not explored. We investigated hypothesis that body mass might decrease detection precancer increase risk even in women undergoing state-of-the-art screening. Methods conducted a retrospective cohort study 944,227 age 30 64 years who underwent cytology human papillomavirus DNA testing (ie, cotesting) at Kaiser Permanente Northern California (January...
Abstract There is limited access to effective cervical cancer screening programs in many resource‐limited settings, resulting continued high burden. Human papillomavirus (HPV) testing increasingly recognized be the preferable primary approach if affordable due superior long‐term reassurance when negative and adaptability self‐sampling. Visual inspection with acetic acid (VIA) an inexpensive but subjective inaccurate method widely used either for or triage of HPV‐positive individuals. A deep...
Abstract The integration of artificial intelligence into clinical workflows requires reliable and robust models. Repeatability is a key attribute model robustness. Ideal repeatable models output predictions without variation during independent tests carried out under similar conditions. However, slight variations, though not ideal, may be unavoidable acceptable in practice. During development evaluation, much attention given to classification performance while repeatability rarely assessed,...
Novel screening and diagnostic tests based on artificial intelligence (AI) image recognition algorithms are proliferating. Some initial reports claim outstanding accuracy followed by disappointing lack of confirmation, including our own early work cervical screening. This is a presentation lessons learned, organized as conceptual step-by-step approach to bridge the gap between creation an AI algorithm clinical efficacy. The first fundamental principle specifying rigorously what designed...
Abstract Background WHO has recommended HPV testing for cervical screening where it is practical and affordable. If used, important to both clarify implement the clinical management of positive results. We estimated performance in Lusaka, Zambia a novel screening/triage approach combining typing with visual assessment assisted by deep-learning called automated evaluation (AVE). Methods In this well-established cancer program nested inside public sector primary care health facilities,...
<b>Objective</b> To assess whether vaccination against human papillomavirus (HPV) increases the risk of miscarriage. <b>Design</b> Pooled analysis two multicentre, phase three masked randomised controlled trials <b>Setting</b> Multicentre in several continents and Costa Rica. <b>Participants</b> 26 130 women aged 15-25 at enrolment; 3599 pregnancies eligible for analysis. <b>Interventions</b> Participants were randomly assigned to receive doses bivalent HPV 16/18 VLP vaccine with AS04...
The effective management of women with human papillomavirus (HPV)-positive, cytology-negative results is critical to the introduction HPV testing into cervical screening. typing has been recommended for colposcopy triage, but it not clear which combinations high-risk types provide clinically useful information. This study included 18,810 Hybrid Capture 2 (HC2)-positive, and who were age ≥30 years from Kaiser Permanente Northern California. median follow-up was 475 days (interquartile range...
HPV testing is more sensitive than cytology for cervical screening. However, to incorporate tests into screening, risk-stratification ("triage") of HPV-positive women needed avoid excessive colposcopy and overtreatment. We prospectively evaluated combinations partial typing (Onclarity, BD) triage, explored whether management could be simplified, based on grouping yielding similar 3-year or 18-month CIN3+ risks. typed ∼9,000 archived specimens, taken at enrollment (2007-2011) the NCI-Kaiser...
Objectives The next round of the American Society for Colposcopy and Cervical Pathology (ASCCP)-sponsored cervical cancer screening management guidelines will recommend clinical actions based on risk, rather than test-based algorithms. This article gives preliminary risk estimates setting, showing combinations 2 most important predictors, human papillomavirus (HPV) status cytology result. Materials Methods Among 1,262,713 women aged 25 to 77 years co-tested with HC2 (Qiagen) at Kaiser...
Abstract We examined whether automated visual evaluation (AVE), a deep learning computer application for cervical cancer screening, can be used on cervix images taken by contemporary smartphone camera. A large number of acquired the commercial MobileODT EVA system were filtered acceptable quality and then 7587 from 3221 women annotated group gynecologic oncologists (so gold standard is an expert impression, not histopathology). tested analyzed multiple random splits using two learning,...
HPV35 has been found in only ∼2% of invasive cervical cancers (ICC) worldwide but up to 10% Sub-Saharan Africa, warranting further investigation and consideration impact on preventive strategies. We studied ethnicity, relation the known steps carcinogenesis, using multiple large epidemiologic studies U.S. internationally. Combining five studies, we measured positivity and, Northern California, observed type-specific population prevalence estimated 5-year risk developing precancer when...
The 2019 ASCCP Risk-Based Management Consensus Guidelines include recommendations for partial human papillomavirus (HPV) genotyping in management of abnormal cervical cancer screening results. guidelines are based on matching estimates intraepithelial neoplasia (CIN) 3+ risk to consensus clinical action thresholds. In support the guidelines, this analysis addresses risks predicted by individual identification HPV 16 and 18.Risk were drawn from a subset women Kaiser Permanente Northern...
The World Health Organization recommends human papillomavirus (HPV) testing for cervical screening. Extended genotyping can identify the highest-risk HPV-positive women. An inexpensive, rapid, mobile isothermal amplification assay (ScreenFire HPV RS test) was recently redesigned to yield four channels ordered by cancer risk (ie, hierarchical approach): HPV16, HPV18/45, HPV31/33/35/52/58 and HPV39/51/56/59/68. Stored specimens from 2076 women (mean age 30.9) enrolled in a colposcopy clinic,...
Visual assessment is currently used for primary screening or triage of screen-positive individuals in cervical cancer programs. Most guidelines recommend and up to at least age 65 years old. We examined images from participants three National Cancer Institute funded studies: ALTS (2864 recruited between 1996 1998) the United States (US), NHS (7548 1993) Costa Rica, Biopsy study (684 2009 2012) US. Specifically, we assessed visibility squamocolumnar junction (SCJ), which susceptible zone...
Cervical cancer is caused by the persistent infection of certain types Human Papillomavirus (HPV) and a leading cause female mortality particularly in low middle-income countries (LMIC). Visual inspection cervix with acetic acid (VIA) commonly used technique cervical screening. While this inexpensive, clinical assessment highly subjective, relatively poor reproducibility has been reported. A deep learning-based algorithm for automatic visual evaluation (AVE) aceto-whitened images was shown...
Cervical cancer is a leading cause of mortality, with approximately 90% the 250,000 deaths per year occurring in low- and middle-income countries (LMIC). Secondary prevention cervical screening involves detecting treating precursor lesions; however, scaling efforts LMIC has been hampered by infrastructure cost constraints. Recent work supported development an artificial intelligence (AI) pipeline on digital images cervix to achieve accurate reliable diagnosis treatable precancerous lesions....