30
$\begingroup$

What are some examples of successful mathematical attempts in clinical setting, specifically at the patient-disease-drug level?

To clarify, by patient-disease-drug level, I mean the mathematical work is approved to be used as part of a decision making process to prescribe a specific treatment for a specific patient?

I do not mean general modeling attempts that study and simulate on a more or less theoretical level.

Additionally, are these mathematical work have to be approved by the regulators like the FDA or a national board of physicians?

$\endgroup$
1
  • 1
    $\begingroup$ Mathematics are everywhere. From producing medical images (for example TC scans) to weight control. The most used decision tree in clinical settings might be "if temperature > 37.5ºC, take some antipyretics". $\endgroup$
    – Pere
    Sep 18, 2019 at 6:21

7 Answers 7

17
$\begingroup$

An example of a simple mathematical/evolutionary game theory model used to determine treatment scheduling in clinical treatment of metastic and castrate resistant prostate cancer can be found at https://www.nature.com/articles/s41467-017-01968-5. While the clinical trial is on-going, initial results show that the model derived treatment schedule offers significant improvement in time to treatment failure when compared to the standard-of-care.

In short, the authors use a 3 population evolutionary game theory model to study the interplay between androgen dependent, androgen independent and androgen producing cancerous cells. They use this model to determine an "adaptive therapy" dosing strategy for abiraterone, a drug that strongly effects androgen dependent cancer cells.Typically, prostate cancer cells develop resistance to abiraterone. Consequently, disease progression and treatment failure is observed in roughly a year (11 months for Prostate Specific Antigen [a biomarker] increase and 16 months for radiographic progression). While small, the study shows that the median time to PSA and radiographic progression is at least 27 months when patients follow the model derived treatment schedule.

Adaptive therapy is designed to take advantage of the "cost of resistance." When abiraterone concentrations are high, it is hypothesized that the relative cost of resistance is lower than the fitness benefit, so cancerous cells are likely to evolve into abiraterone resistant strains. Adaptive therapy aims to use treatment holidays to protect against the development of these resistant strains. In general, treatment is suspended when PSA reaches half of pre-treatment levels, and only restarts once pre-treatment PSA levels have been reached.

$\endgroup$
24
$\begingroup$

Machine learning is on its way to provide the type of personalized health care referred to by the OP. In June of this year the FDA (US Food and Drug Administration) has proposed a regulatory framework for machine learning algorithms in medical decision making.

The document distinguishes "locked algorithms" (basically decision trees) and "adaptive algorithms" (a supervised learning algorithm that will "change its behavior using a defined learning process", as the inputs are given new data). Only the locked algorithms are currently approved by the FDA, and the document tries to set up a pathway for supervised machine learning algorithms to be used in health care.

$\endgroup$
14
$\begingroup$

I'm going to conflate mathematics with statistics as Carlo Beenakker did. Then the earliest application that I know of is that Decision Trees were invented by Breiman et al. to analyze the issue of- among people who have had a heart attack, which patients were most likely to have another heart attack. I also believe that Ed Frenkel, in his autobiography, claimed to have developed a similar methodology to explain to doctors how to triage. A really beautiful application is something called James-Stein estimation which deals with the inadmissability of the mle for estimating (true) means of many variable. The short story is that if you have 4 or more series of observations, then the individual estimated mean of each series of observations is not the best estimates of the real means. This was applied by Efron (mentioned in various books and papers including "Large Scale Inference") to the question of deciding which genes (via gene expression) were likely to be influential in causing a specific cancer.

$\endgroup$
11
$\begingroup$

From the Automatic Control Laboratory at ETH Zürich, a project on automating anaesthesia:

The first steps to introduce feedback control in anesthesia were undertaken more than ten years ago. The project encompasses all control system aspects, namely

  • design of experiments and open loop identification

  • modeling of biological complex systems

  • statistical analysis of large population data

  • control design and simulation

  • pilot studies and clinical routine validation

Automated Control of Anesthesia was successfully tested during surgery on more than 200 patients and 650 hours.

The successful application of our ideas is made possible by the close cooperation with: the Department of Anaesthesiology at the University Hospital in Berne (Inselspital), where volunteers are enrolled for model identification and controllers are validated in clinical studies.


Papers:


Theses:

$\endgroup$
10
$\begingroup$

I know of one mathematical system that was used before the widespread use of CT scans of brain in diagnosing stroke type. That of Scoring methods, which gave at those days a clinical decision that is by way much better than the clinical decision made by highly ranked experienced professional Consultant Neurologists. An example of a score that I myself researched was the Allen Score in differentiating stroke type, i.e. hemorrhagic from ischemic strokes.

I personally think that along similar lines we can get machines to overcome the experienced human diagnosis in many fields in medicine, albeit not always of course.

$\endgroup$
4
$\begingroup$

Amazingly, noone mentionned the use of the inverse Radon transform in the scanner / medical imaging.

$\endgroup$
2
$\begingroup$

The APGAR test. It may not be the most interesting example because it's maths are apparently just a sum but in fact it's a simple decision tree. However, it's approved, useful and widely used.

$\endgroup$
3
  • 1
    $\begingroup$ I would hardly call summing five integers between 0 and 2 "math"... $\endgroup$ Jun 3, 2020 at 7:31
  • 1
    $\begingroup$ I agree that using the test is very simple math (in fact, that is already acknowledged in the answer), but tuning and validating it involves more interesting math. Decision trees are an interesting part of statistics - unless you say they aren't math because they are data science. $\endgroup$
    – Pere
    Jun 3, 2020 at 8:18
  • $\begingroup$ Well, here's what I tried to say: I don't consider it "very simple math", because I don't consider it math at all. $\endgroup$ Jun 3, 2020 at 9:19

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.