Jeffrey P Bond and Scott D Luria

An old man, and pale: anemia. Also thin: first thought, cancer. Second thought, tuberculosis, alcoholism, some other chronic process . . . the computer provided him with a differential, complete with probabilities of diagnosis. From The Andromeda Strain (1969),1 by Michael Crichton

How is cancer information exchanged among laypersons, clinical professionals, and medical researchers? High hopes for the role of computers in medical information exchange have been reflected in science fiction for decades. After at least two information technology paradigm shifts (personal computers and the Internet) and countless successful implementations of, for example, shared electronic records, knowledge bases, decision support systems, speech-to-text tools, natural language processing tools, or remote monitoring devices in a variety of medical and nonmedical settings, we are in the process of realizing these high hopes. An important component of cancer information exchange infrastructure is support for the use of knowledge derived from medical research in combination with patient data to guide cancer-related decisions, that is, support for evidence-based cancer medicine, the central focus of this chapter.

We focus on three classes of scenarios that require informatics support: a healthcare professional making a cancer-related decision about a patient, someone (referred to here as a layperson) making a cancer-related decision without medical training and without consulting a healthcare professional, and a researcher making a decision regarding a cancer-related experiment. Each such decision involves a collection of potential actions, information that describes the context of the action, and knowledge derived from scientific experiments that constitutes evidence bearing the merit of the potential actions given the context (Figure 10.1).

The purpose of this chapter is to describe how clinicians, patients, and researchers can find information related to cancer prevention, diagnosis, treatment, and research on the World Wide Web. (Although WWW resources will not necessarily behave as they did at the time this chapter was written, we think it is important to include numerous specific examples.) Here infrastructure, then, refers to standards, software, and digital information rather than hardware. We emphasize digital media but do not intend to imply that other media (such as printed textbooks, classroom lectures, handwritten notes, pamphlets, posters in subways, radio/television spots, or phone support lines) cannot be preferable to computer-related infrastructure for accomplishing particular goals. Currently the support for integration of knowledge bases and context data available to most clinicians is poor; in nearly all examples below, this integration is manual.

We do not attempt comprehensive coverage of medical informatics (for example, electronic medical records,2 HIPAA,3 or telemedicine,4), which includes cancer informatics. Readers interested in resources describing broader aspects of bioinformatics or medical informatics might consult one of the more comprehensive resources listed in Table 10.1. Readers interested in a variety of online bioinformatics databases and software might consider one of the URLs given in Table 10.2.

We focus on two structures that facilitate information exchange: indexes and graphs, analogous to the index and table of contents of a book or the index accessed using a search box and the directory of a Web resource. Given a set of objects, indexes associate words or phrases with subsets, which can be combined using set algebra. For example, one can search for clinical trials at http://clinicaltrials.gov with a query such as disease="breast cancer", experimental treat-ment="surgery", and state="Vermont." Graphs consist of nodes and relationships between nodes. In the cases described here, the nodes are Web pages and the relationships are represented by links or buttons. An example of such a graph is a collection of Web pages including directory pages that support browsing. Graphs need not be hierarchical like a physical filing system because there may be multiple paths between Web pages.

One of the primary difficulties in developing informatics support for cancer-related decisions results from the level of detail we expect. Informatics support for book purchases necessarily includes rudimentary support for cancer informatics in that it serves to identify books on cancer. Searches based on author, title, or keywords from a simple vocabulary (for

FIGURE 10.1. Schematic describing how potential actions are based on and inform knowledge derived from scientific experiments and the information that describes the context of the action.

example, breast cancer) often serve to identify sets of book records that can either be browsed manually or productively ranked (for example, the number of times a book has been purchased is evidence related to the quality of the book). We expect cancer informatics support to do better than simply finding books that will answer questions, we expect to identity chapters or paragraphs that answer our questions. Ideally, informatics resources should answer questions directly.

Such fine granularity requires, relative to the task of simply identifying of books on cancer, a complex vocabulary and a query syntax that supports context data. In this chapter, we focus on four components of such a solution: vocabulary resources that support intelligent indexing, manually created graphs for browsing, manual indexing of documents based on their generalilty and usefulness, and integration of the search with patient data.

Laypersons' Decisions

A variety of cancer information portals and sources of information exist on the World Wide Web for laypersons (Table 10.3). We focus on two comprehensive resources: Cancer.gov, which is maintained by the U.S. National Cancer Institute (NCI), and Cancer.org, which is maintained by the American Cancer Society (ACS).

Cancer.gov Features

Cancer.gov is a source of general cancer information as well as information on NCI research programs, NCI research funding, cancer statistics, and clinical trials. The cancer information section, http://cancer.gov/cancerinfo, includes a variety of information that is intended for laypersons seeking information about cancer etiology, diagnosis, and treatment. Below we highlight certain features of Cancer.gov.

"What You Need to Know About Cancer" (WYNTK) documents. http://cancer.gov/cancerinfo/wyntk provides a starting point for obtaining information about cancer in general and more than 20 specific cancers. Each such document includes information about symptoms, diagnosis, and treatment. An overview document provides a general and readable introduction to cancer with pointers to other NCI resources.

Dictionary. http://cancer.gov/dictionary/ provides nontechnical definitions of cancer-related terms. There are currently more than 4,000 terms, with approximatly 40 terms added each month. For example:

metastasis (meh-TAS-ta-sis) The spread of cancer from one part of the body to another. A tumor formed by cells that have spread is called a metastatic tumor or a metastasis. The metastatic tumor contains cells that are like those in the original (primary) tumor. The plural form of metastasis is metastases (meh-TAS-ta-seez).

Physician Data Query (PDQ). http://cancer.gov/cancerinfo/ pdq is "NCI's comprehensive cancer database, containing peer-reviewed summaries on cancer treatment, screening, prevention, genetics, and supportive care" and also provides information on clinical trials and directories. Although most of this information is directed at healthcare professionals, many of the topics have links to pages written specifically for patients and linked to dictionary entries. Compared to the WYNTK documents, "PDQ patient" documents are more specific; for example, there is a single WYNTK document on

TABLE 10.1. General biomedical informatics

resources.

Resource

Scope

Bioinformatics: A Practical Guide to the

Sequences, structures, genomes, alignments,

Analysis of Genes and Proteins53

phylogenetic trees

Bioinformatics Sequence and Genome

Sequences, structures, genomes, alignments,

Analysis54

phylogenetic trees

NCBI Handbook23

NCBI databases (sequence, structure, journal

Was this article helpful?

0 0
10 Ways To Fight Off Cancer

10 Ways To Fight Off Cancer

Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.

Get My Free Ebook


Post a comment