Abstract

Cancer screening programs have been implemented in many different countries for many years to collect information of the fatal diseases, to provide early diagnosis, to support medical research, and to help governments making policies. However, few of those programs have utilized latest data science technologies, therefore not be able to generate the maximum benefits from those programs. To overcome this problem and improve the quality of cancer screening programs, this paper firstly (i) reviews the typical architecture and IT technologies used in current screening programs and recognizes their limitations; then (ii) introduces recent developments in data science that could be implemented in screening programs; finally (iii) proposes the structure of General Medical Screening Framework (GMSF), which could be developed to host future cancer screening programs that will advance data coverage, data accuracy, data usage and lower in the costs. The structure of GMSF and its key elements are described in this paper and some practical approaches to build GMSF are discussed. This work might initialize a series or research to bring the latest IT technologies, particularly data science technologies, into cancer screening programs, and significantly increase the efficiency and reduce the cost of future cancer screening programs.

Highlights

  • Due to the high mortality of cancer[1] (16.0/million in urban China[2]), many cancer screening programs have been implemented in different regions through different medical organizations[3,4,5]

  • This paper proposes a General Medical Screening Framework (GMSF) with the core of General Medical Screening System (GMSS) that would arguably help to design and implement new cancer screening programs with higher quality and more economic and scientific benefits

  • Considering current situation of cancer screening programs and the potential development in the future, this paper proposes a general medical screen framework that aims to be an embryonic concept for future cancer screening programs

Read more

Summary

Introduction

Due to the high mortality of cancer[1] (16.0/million in urban China[2]), many cancer screening programs have been implemented in different regions through different medical organizations[3,4,5]. The Cancer Screening Program in Urban China (CanSPUC) has been implemented in 14 provinces to provide cancer screening for lung, breast, colorectal, esophageal, gastric and liver cancers[6] from 2012. The benefits of those cancer screening programs have been recognized, and the cost-utility analysis has been conducted on some programs to justify the value and help governments to make decisions to provide most cost-effective screening programs[7]. In order to achieve that, this paper firstly reviews the architecture and techniques used in many current cancer screening programs and shows their problems and limitations, and briefly introduces some latest development in data.

A typical structure of existing screening programs and its limitations
Latest development of data science technologies and software engineering
Data volume
Data structure
Data lifecycle
Data collection
Data storage
Data access
Data verification and reasoning
Software engineering and software tools
New roles
Vision and goals
Data science technology
Professional groups
Program coverage
Summary
Discussion
Feasibility
Practical approaches
Legacy systems
Conclusions and further research topics
Declarations
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call