Editorials |

Sleep Tools FREE TO VIEW

Max Hirshkowitz, PhD; Amir Sharafkhaneh, MD
Author and Funding Information

Affiliations: Houston, TX 
 ,  Dr. Hirshkowitz is Associate Professor, Department of Medicine, Department of Psychiatry, and Dr. Sharafkhaneh is Assistant Professor, Department of Medicine, Veterans Affairs Medical Center and Baylor College of Medicine, Sleep Disorders and Research Center.

Correspondence to: Max Hirshkowitz, PhD, Sleep Disorders and Research Center, VA Medical Center 116A, 2002 Holcombe Blvd, Houston, TX 77030-4298; e-mail: maxh@bcm.tmc.edu

Chest. 2001;119(5):1303. doi:10.1378/chest.119.5.1303
Text Size: A A A
Published online

Ralph Waldo Emerson once wrote, “All the tools and engines on earth are only extensions of man’s limbs and senses.”1 Tools born from the marriage of health and computer sciences have tremendously extended both our reach and sight. In sleep medicine, computerization has long boasted great promise. That promise is at long last reaching fruition. Perhaps the greatest attraction of computerization in sleep medicine relates to data reduction. From the very beginning, sleep specialists had to devise rules in order to summarize the hours of data and miles of paper (now megabytes of disk files). On average, it takes a well-trained polysomnographic technologist and polysomnographer several hours to distill a sleep recording to the essence required for clinical purposes. As sleep medicine grows, so does the pressure to develop more efficient tools for recording and reducing data.

As with any tool, computerized sleep recorder-analyzers present both advantages and disadvantages. Automation speeds the process, thereby creating efficiencies. However, automation often removes us further from the actual data. During development, these tools are, in engineering terms, “benchmarked.” This may be accomplished by using specific “test data sets” or specific, well-defined test sample problems. In evidence-based medicine lingo, these would represent the sample on which a system is validated. Furthermore, the sample composition and characteristics define the population to which the results may be generalized. For example, the sensitivity and specificity for detecting obstructive sleep apnea syndrome (OSAS) may differ when a system is tested on young adults compared to elderly, patients with heart disease compared to otherwise healthy individuals, or patients with insomnia compared to patients with excessive sleepiness. Often, differences are subtle, making little difference in outcome; however, in the clinic we do not play the odds. Accurate information is needed in each individual case to ensure that treatment decisions are made rationally. Thus, it is essential for the craftsman to understand the limitations of his or her tools.

In this issue of CHEST (see page 1387), Cirignotta and colleagues provide an example of a computerized sleep system shortfall. MESAM (MAP; Martinsried, Germany) is a widely used, well-engineered cardiopulmonary sleep recorder-analyzer. It can detect, with relative accuracy, desaturation events in patients afflicted with OSAS who do not have clinically significant cardiopulmonary disease (ie, the “benchmark” is noncomplicated OSAS). Cirignotta and colleagues provide examples where accuracy is compromised—specifically, when the device is used to evaluate what they call “complicated” OSAS. Furthermore, the authors probe the underlying polysomnographic features that interact with the detection algorithm producing this problem. It is precisely this type of approach that provides useful guidance to further develop the tool. It also highlights the need for benchmark test data sets to include “complicated” OSAS. This is especially the case for computerized polysomnographic and sleep cardiopulmonary recorder-analyzer validation because these tools are gaining popularity and becoming more widely used by individuals with less expertise.

In summary, computerized polysomnographic and sleep cardiopulmonary equipment can be a very effective tool and is extending the reach of sleep medicine. However, it has shortcomings. As with any tool, a skilled user must understand how that tool functions and be aware of its limitations. Recognizing the limitations of a tool is often the first step to developing a better tool. Not being aware of the limitations of a tool can lead to its misapplication, thereby posing a hazard. As Czech novelist Milan Kundera wrote, “A worker may be the hammer’s master, but the hammer still prevails. A tool knows exactly how it is meant to be handled, while the user of the tool can only have an approximate idea.”2


Emerson, RW (1968)Society and solitude. AMS Press. New York, NY:
Kundera, M. Book of laughter and forgetting. 1980; A.A. Knopf. New York, NY:.




Emerson, RW (1968)Society and solitude. AMS Press. New York, NY:
Kundera, M. Book of laughter and forgetting. 1980; A.A. Knopf. New York, NY:.
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Find Similar Articles
CHEST Journal Articles
PubMed Articles
  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543