Pintea, C.-M., and Palade, V. A glass-box interactive machine learning approach for solving NP-hard problems with the human-in-the-loop; https://arxiv. org/abs/1708.01104

8. Holzinger, A., Schantl, J., Schroettner, M., Seifert, C., and Verspoor, K. Biomedical text mining: State-of-the-art, open problems and future challenges. Chapter in Interactive Knowledge Discovery and Data Mining Biomedical Informatics, Lecture Notes in Computer Science LNCS 8401, A. Holzinger and I. Jurisica, Eds.

Springer, Berlin, Heidelberg, Germany, 2014, 271–300.

9. Islam, A., Milios, E., and Keselj, V. Text similarity using Google tri-grams. In Proceedings of the 25th Canadian Conference on Artificial Intelligence (Toronto, Canada, May 28–30). Springer, Toronto, Canada, 2012, 312–317.

10. Kintsch, W. Predication. Cognitive Science 25, 2 (2001),

173–202.

11. Landauer, T. K. and Dumais, S. T. A solution to Plato’s problem: The latent semantic analysis theory of acquisition, induction, and representation of knowledge.

Psychological Review 104, 2 (1997), 211–240.

12. Landauer, T.K., Foltz, P. W., and Laham, D. An introduction to latent semantic analysis. Discourse Processes 25, 2 and 3 (1998), 259–284.

13. Landauer, T.K., Laham, D., and Derr, M. From paragraph to graph: Latent semantic analysis for information visualization. Proceedings of the National Academy of Sciences 101, 1 (Apr. 6, 2004), 5214–5219.

14. Larsen, K. R. and Bong, C. H. A tool for addressing construct identity in literature reviews and meta-analyses. MIS Quarterly 40, 3 (Sept. 2016), 529–551.

15. Larsen, K.R., Michie, S., Hekler, E.B., Gibson, B., Spruijt-Metz, D., Ahern, D., Cole-Lewis, H., Ellis, R.J.B., Hesse, B., Moser, R.P., and Yi, J. Behavior change interventions: The potential of ontologies for advancing science and practice. Journal of Behavioral Medicine 40, 1 (Feb. 2017), 6–22.

16. Valle-Lisboa, J. C. and Mizraji, E. The uncovering of hidden structures by latent semantic analysis.

Information Sciences 177, 19 (Oct. 2007), 4122–4147.

17. Zhiwei, H., Meiping, C., Jimin, W., Qing, S., Chao, Y., Xing, D., and Zhonggao, W. Improved control of hypertension following laparoscopic fundoplication for gastroesophageal reflux disease. Frontiers of Medicine

11, 1 (Mar. 2017), 68–73.

 

David Gefen ( [e-mail]) is a professor in the Decision Sciences and MIS Department, Academic Director of the Doctorate in Business Administration Program, and Provost Distinguished Research Professor in the LeBow College of Business at Drexel University, Philadelphia, PA, USA.

Jake Miller ( [e-mail]) is an assistant clinical professor in the Management Department in the LeBow College of Business at Drexel University, Philadelphia, PA, USA.

Johnathon Kyle Armstrong ( Kyle.Armstrong@ibx. com) is a research scientist at Independence Blue Cross, Philadelphia, PA, USA.

Frances H. Cornelius ( [e-mail]) is a professor in the College of Nursing and Health Professions and Chair of the MSN Advanced Role Department, Complementary and Integrative Health Department, and coordinator of Clinical Nursing Informatics Education in the College of Nursing and Health Professions at Drexel University, Philadelphia, PA, USA.

Noreen Robertson ( [e-mail]) is the Associate Vice Dean for research at Drexel University College of Medicine and a research assistant professor in the Department of Biochemistry & Molecular Biology at Drexel University, Philadelphia, PA, USA.

Aaron Smith-McLallen ( [e-mail]) is the Director of Data Science and Health Care Analytics at Independence Blue Cross, Philadelphia, PA, USA.

Jennifer A. Taylor ( [e-mail]) is an associate professor of environmental and occupational health in the School of Public Health at Drexel University, Philadelphia, PA, USA.

This study was supported by Drexel Grant #282847.

Copyright held by authors.

Publication rights licensed to ACM. $15.00

dicative of the potential in applying LSA to such contexts. We created a semantic model that identified known relationships among medical terms, relating diagnoses and treatments. We scrubbed the sample of most demographic data; the dataset was too small anyway to allow cross-sectional analysis. Constructing a model from a larger, more detailed dataset could yield substantial potential for medical discovery. Comparing reports across patients could provide even more information, as by, say, enabling creation of a “typical” profile of care/ treatment trajectory, as well as diagnosis and prognosis as they apply to disease, condition names, or ICD codes. Such a profile could conceivably lead to early detection and allow identifying exceptional cases in need of immediate medical attention.

A typical profile for a condition or ICD code could help create a method to at least partly support phase IV testing of new drugs involving long-term monitoring of the effects of drugs following approval by the U.S. Food & Drug Administration. LSA could improve this process not only by automating it but also by identifying a drug’s possible indirect effects, or the effects associated with the drug but only through other diagnosis. So, for example, if drug A is associated with condition B and condition B is associated with condition C, then LSA will identify that A and C might be related. A human examiner might not notice it but could be aided by LSA to identify possible connections of interest for the expert to consider; see Holzinger et al. 7 for more on interactive machine learning.

Analyzing medical records could also allow comparison of diagnosis and prognosis across populations (such as differentiating between men and women). Accounting for demographics could also indicate the prevalence of diagnosis and prognosis by age and by geographical area, possibly indicating hazardous environmental conditions.

Moreover, given the diversity within so-
ciety, running LSA on medical records
could also allow quasi-experimental de-
sign studies, as in, say, comparing clin-
ics in areas where unique treatments
are allowed against those where they
are not. Planning such an experiment
would be difficult and IRB approval
might not always be forthcoming, but
if the treatment conditions occur in the
population, then studying them would
not be so contentious and could be-
come routine. LSA could support such
an after-the-effect examination.

Combining LSA cosine distances with TDM frequency values may also allow identification of extraordinary cases that are closely related to a condition but very rare. In the data we had, we omitted terms that appeared in fewer than four records, but the relationships between rare diseases and more common ones might suggest new avenues of research for a range of health issues. As a supplement to medical practice, a text-analytic approach might be able to suggest alternative diagnoses based on documented symptoms that might otherwise be attributed to more common conditions.

Above all, a key advantage of LSA is that it allows rank ordering of related terms. Being able to assign numbers, and hence categorize how much a condition is related to other conditions, could provide insight about identifying what symptoms, and how much more than others, they might indicate a problem (such as hypertension).

Conclusion

In a musical parody, lyricist David Lazar wrote in a song called “Dr. Freud” that Sigmund Freud’s disciples said, “…by God, there’s gold in them thar ills.” There certainly was. Maybe there also is in gleaning medical insight from medical records documents through LSA.

 

References

1. Beel, J., Gipp, B., Langer, S., and Breitinger, C. Research-paper recommender systems: A literature survey. International Journal on Digital Libraries 17, 4 (Nov. 2016), 305–338.

2. Deerwester, S., Dumais, S. T., Furnas, G. W., Landauer, T.K., and Harshman, R. Indexing by latent semantic analysis. Journal of the American Society for Information Science 41, 6 (1990), 391–407.

3. Evangelopoulos, N., Zhang, X., and Prybutok, V.R. Latent semantic analysis: Five methodological recommendations. European Journal of Information Systems 21, 1 (Jan. 2012), 70–86.

4. Gefen, D., Endicott, J., Fresneda, J., Miller, J., and Larsen, K.R. A guide to text analysis with latent semantic analysis in R with annotated code studying online reviews and the Stack Exchange community. Communications of the Association for Information Systems 41, 21 (Dec. 2017), 450–496.

5. Gefen, D. and Larsen, K. Controlling for lexical closeness in survey research: A demonstration on the technology acceptance model. Journal of the Association for Information Systems 18, 10 (Oct. 2017), 727–757.

6. Gomez, J.C., Boiy, E., and Moens, M.-F. Highly discriminative statistical features for email classification. Knowledge Information Systems 31, 1 (Apr. 2012), 23–53.

7. Holzinger, A., Plass, M., Holzinger, K., Crisan, G.C.,