Cervical Cancer Projects

Pocket Colposcope Design, Development and Testing

Project lead: Nimmi Ramanujam, PhD 

Robert W. Carr Jr., Professor of Biomedical Engineering
Professor of Pharmacology and Cancer Biology
Professor of Global Health
Director, Center for Global Women's Health Technologies

Project lead: Jenna Mueller, PhD

Post-Doctoral Researcher
Biomedical Engineering

Pratt School of Engineering

We have developed a portable, low cost, transvaginal POCkeT Colposcope for the screening of cervical cancer (Figure 1). Our POCkeT Colposcope can bring colposcopy to places previously limited by VIA and improve accessibility, quality control and specificity of cervical cancer screening. Image concordance analysis was completed by eight international physicians and concluded that performance of our POCkeT Colposcope was equivalent to the more expensive standard-of-care colposcope at identifying precancerous lesions of the cervix (our colposcope device and standard-of-care colposcope as shown in Figure 2). 

Figure 1: A working prototype of our POCkeT Colposcope interfaced with a Google Nexus 5 Android Smartphone and displaying a HSIL (high grade lesion). Note the US Penny shown for scale which has a diameter of 0.75 in (19.05 mm). The TVDC (A) can interface with laptops, smartphones, and tablets to store images captured through a USB 2.0 wired connection (1), a manual focus knob allows for coarse adjustment of the focus before the integrated autofocus mechanism performs during burst image capture (2) allows for imaging from 10 to 50 mm from the tip of the device to the cervix (3).

Figure 2: (A-C) Representative cervical images with the application of acetic acid (vinegar) captured at 3.75 X with Leisegang Optik 2 Digital Colposcope (18 megapixel color CMOS detector) at a working distance of 300 mm and (D-F) concordant images at 4X with 5.0MP POCkeT Colposcope at a working distance of 40 mm. Normal Cervix (AD), biopsy confirmed LSIL/CIN 1 (BE) at 6 o’clock position, and biopsy confirmed HSIL/CIN 3 pathology (CF) confirmed at the 5 and 11 o’clock.

Because our device is inserted transvaginally, the ability for digital image capture is brought closer to the cervix (35 mm), therefore decreasing the resolution required (5.0 MP) to capture a highly magnified image. Our device was modeled after a tampon, with the eventual goal to have women self-insert our device to capture an image of their cervix, for later physician review. Our device has internal LEDs for both white and green light illuminated image capture. Image concordance analysis was completed by eight international physicians and concluded that performance of our POCkeT Colposcope was equivalent to the more expensive standard-of-care colposcope at identifying precancerous lesions of the cervix. Paired cervical images (our colposcope device and standard of care colposcope as shown in Figure 2), with biopsy confirmed pathologies, were randomized and blinded by device prior to review. Implications of these findings have the potential to create increased access to a culturally appropriate screening technology, thus reducing the burden of cervical cancer throughout the developing world.

Clinical Implementation and Validation of the Pocket Colposcope in Local and International Settings

Project lead: Marlee Krieger, MS 

Research Program Administrator, Biomedical Engineering 
Program Coordinator, Center for Global Women's Health Technologies

Project lead: John Schmitt, MD 

Professor of Obstetrics and Gynecology 
Chief, Benign Gynecology
 

Dr. Schmitt leads our international image concordance study where eight physicians around the globe assess the level of agreement between images collected with the POCkeT Colposcope and that of a state-of-the-art colposcope. Our analysis indicates a high level of agreement between the two systems at identifying precancerous lesions of the cervix (1). The map below shows countries collaborating on the image concordance study.

Duke University Medical Center (DUMC): Durham, North Carolina - Jennifer Gallagher
Kilimanjaro Christine Medical Center (KCMC): Moshi, Tanzania - Bariki Mchome, MD,  Olala Oneko, MD
La Liga Peruana de Lucha contra el Cáncer: Lima, Peru - Gino Venegas, MD
Chennai Cancer Institute (WIA): Chennai, India - J.S Malliga, MD
University of Nairobi: Nairobi, Kenya
Kenyatta University: Kenya - Anthony Wanyoro, MD
Mombasa, Kenya: Jennifer Smith, PhD.
University of Zambia: Lusaka, Zambia - Groesback Parham, MD

mHealth Technology for Cervical Cancer Screening

Project lead: Lavanya Vasudevan, PhD 

Research Scholar
Duke Global Health Institute

The key roadblocks before implementing the POCkeT Colposcope on-site in developing countries are (1) image interpretation and (2) access to healthcare to reduce distance and communication barriers. The development of a mobile health (mHealth) application will help breakdown these barriers and thus set the stage for follow-on funding for clinical investigations with the mHealth integrated platform. Key informant observation focus groups and interviews have been used to assess the feasibility of and guide the design and development of a prototype mobile health (mHealth) application using open source Android software and smartphone/tablet hardware that integrates: point of care diagnostic/screening, electronic health records, and decision support in the use of the POCkeT Colposcope. Baseline data on mobile phone ownership and use were collected from 16 health providers (physicians, residents and nurses) from Kilimanjaro Christian Medical Centre in Moshi, Tanzania and analyzed. Results suggest high ownership of smartphones in this population as well as ability to use key phone features to perform tasks such as image acquisition, email, and file attachments that are relevant to the mHealth application. Acceptability of using mobile phones to aid cervical cancer screening was also high in this population. Based on this user data, we are in the process of developing an integrated mHealth software solution to enhance the utility of the POCkeT Colposcope.

Evaluating Technologies from Provider, Patient and Administrator Perspectives to Optimize Uptake

Project lead: Rae Jean Proeschold-Bell, PhD 

Associate Research Professor
Director Evidence Lab
Duke Global Health Institute

GWHT incorporates process evaluation and implementation science as part of its larger multidisciplinary approach to the design of health technologies for women. The team at GWHT includes Rae Jean Proeschold-Bell, PhD (Figure 4) a psychologist who is expert in the design and evaluation of health interventions. Some of the evaluation activities of GWHT to date are:

1) Data collection on the experience of using the POCKeT Colposcope, from the perspectives of medical providers, clinic administrators, and patients;

2) Documenting the iterative process of technology design, as user feedback leads to technology improvements multiple times; and

3) Comprehensive understanding of the context in which technologies are used and how local structures (e.g., clinic infrastructure and staff) and beliefs influence the uptake of health technologies.

Global Value Chain Analysis

Project lead: Danny Hamrick, MIS

Research Associate
Center on Globalization, Governance & Competitiveness

Global value chain (GVC) analysis is a methodology used to help researchers and policy makers better understand the ecosystem in which health commodities operate by providing a holistic approach to examining the various actors involved in the development, production, distribution and retailing of healthcare commodities and services, policies that affect healthcare decisions and adoptions of new practices, as well as the linkages of actors and activities. Understanding the ecosystem in which a health commodity operates yields important insights into factors that influence scaling in healthcare including: governance structures (who sets the ‘rules of the game’), policies, flows and allocation of information and resources and the organization of actors involved in the health commodity and the power relationships among actors. The methodology examines the variables that drive change, including customer segments, distribution channels, and regulatory environment. It also considers carefully the critical factors that influence final outcomes. The framework is inherently generalizable with insights regarding scalability easily transferred across countries and regions. GVC allows for the identification of crucial leverage points- parts of the chain where strategic interventions can have transformative effects along the entire chain and influence the success of scaling efforts. Leverage points can be seen as both actors in the chain and specific segments in the chain, such as distribution. By better understanding the chain and identifying potential leverage points early on, implementers can make strategic decisions to help increase the possibility of successful program launches and scaling efforts.

Human Centered Design, Development and Manufacturing

Project lead: Robert Miros, BSE, BA

Product Design
3rd Stone Design

Product development efforts for the POCkeT Colposcope began with comprehensive need-finding, quantitative research review, and key informant interviews. From these insights, our group formulated a product requirements specification to define core design parameters. Concept sketches, user interface mock-ups, and hard and soft models were created to help hone in on the most viable systems to pursue for prototype development. These subsequent prototypes were rapidly iterated using a combination of 3D Computer Aided Design (CAD) software, PolyJet and Fused Deposition Modeling (FDM) 3D printing, and high-resolution Computer Numerical Controlled (CNC) milling. Each prototype system is empirically evaluated using international industry standards from ISO, IEC, ANSI, among others. Human factors analysis on each prototype is conducted with the help of our clinical partners in local and international settings.

The POCkeT Colposcope incorporates a custom microprocessor-driven system into the hardware components to optimize the ease of use and implementation. The complete system can interface with laptops, smartphones, and tablets to store images captured through a USB 2.o wired connection. LED adjustment and image capture are both controlled on the handheld device. A manual focus slider allows for coarse adjustment in addition to an integrated fine autofocus mechanism during burst image capture. This focus slider allows the user to capture detailed images at working distances (from the tip of the device to the cervix) of 10 to 35 mm. Currently, we are actively collaborating with industry partners on further product development and transfer to manufacturing of the POCkeT Colposcope.

“Smart Health” algorithm to automate cervical cancer screening

Project lead: Guillermo Sapiro, PhD 

Professor 
Electrical and Computer Engineering 
Biomedical Engineering

The goal of this program is to develop a “Smart Health” algorithm to automate cervical cancer screening to allow minimally trained community health providers to use physician intelligence or perhaps surpass physician intelligence at the point-of-care setting.  Being able to rapidly and effectively identify patients at highest risk of cervical cancer with our POCkeT colposcope and smart health algorithms is expected to allow for rapid risk assessment and appropriate follow up care which if successful, will reduce the morbidity and mortality associated with this disease. 

Commercialization of the Pocket Colposcope

Project lead: Rob Hallford, MBA 

Assistant Director 
Duke Office of Licensing and Ventures

Project lead: Shweta Krishnan, PhD. 

Licensing and Market Research Analyst 
Duke Office of Licensing and Ventures

The Duke Office of Licensing and Ventures (OLV) is the licensing and venture creation arm of Duke University and the Medical Center. The office aims to lead efforts for faculty inventors to commercialize their technology by protecting intellectual property, enabling entrepreneurial efforts, and industrial partnerships. 

As part of our efforts to commercialize the POCKeT Colposcope, we are constantly working with Dr. Ramanujam and her team to patent and thereby protect novel ideas for the device. We are conducting market research to identify applications for this device outside of cervical cancer screening, which will take us to a target product profile and maximize the breadth of POCKeT Colposcope users. Our goal will then be to reach out to potential licensees and foster industrial collaborations, which will aid in the clinical development, regulatory filings and commercialization of this product.

 


 

References:
1. Mueller J*, Asma B*, Lam C ,Krieger M, Asiedu M, Gallagher J, Hariprasad R, Malliga J, Mchome B, Oneko O, Taylor P, Venegas G, Wanyoro A, Muasher L, Schmitt J, Ramanujam N. Evaluation of a point of care tampon (POCkeT) colposcope as a screening device for cervical cancer among international physicians. Journal of Lower Genital Tract Disease, 2016. Submitted.

*Authors contributed equally