Now Swedish researchers at Uppsala University are revealing that this methodology is worthless when it comes to practical problems. The article is published in the journal Pattern Recognition Letters.
Today there is rapidly growing interest in ‘intelligent’ computer-based methods that use various classes of measurement signals, from different patient samples, for instance, to create a model for classifying new observations. This type of method is the basis for many technical applications, such as recognition of human speech, images, and fingerprints, and is now also beginning to attract new fields such as health care.
“Especially in applications in which faulty classification decisions can lead to catastrophic consequences, such as choosing the wrong form of therapy for treating cancer, it is extremely important to be able to make a reliable estimate of the performance of the classification model,” explains Mats Gustafsson, Professor of signal processing and medical bioinformatics at Uppsala University, who co-directed the new study together with Associate Professor Anders Isaksson.
To evaluate the performance of a classification model, one normally tests it on a number of trial examples that have never been involved in the design of the model. Unfortunately there are seldom tens of thousands of test examples available for this type of evaluation. In biomedicine, for instance, it is often expensive and difficult to collect the patient samples needed, especially if one wishes to analyze a rare disease. To solve this problem, many different methods have been proposed. Since the 1980s two methods have completely dominated research, namely, cross validation and resampling/bootstrapping.
“This has entailed that the performance assessment of virtually all new methods and applications reported in the scientific literature in the last 25 years has been carried out using one of these two methods,” says Mats Gustafsson.
In the new study, the Uppsala researchers use both theory and convincing computer simulations to show that this methodology is worthless in practice when the total number of examples is small in relation to the natural variation that exists among different observations. What is considered a small number depends in turn on what problem is being studied-in other words, it is impossible to determine whether the number of examples is sufficient.
“Our main conclusion is that this methodology cannot be depended on at all, and that it therefore needs to be immediately replaces by Bayesian methods, for example, which can deliver reliable measures of the uncertainty that exists. Only then will multivariate analyses be in any position to be adopted in such critical applications as health care,” says Mats Gustafsson.
Mats Gustafsson | alfa
Earthquake researchers finalists for supercomputing prize
19.11.2018 | University of Tokyo
Putting food-safety detection in the hands of consumers
15.11.2018 | Massachusetts Institute of Technology
Researchers at the University of New Hampshire have captured a difficult-to-view singular event involving "magnetic reconnection"--the process by which sparse particles and energy around Earth collide producing a quick but mighty explosion--in the Earth's magnetotail, the magnetic environment that trails behind the planet.
Magnetic reconnection has remained a bit of a mystery to scientists. They know it exists and have documented the effects that the energy explosions can...
Biochips have been developed at TU Wien (Vienna), on which tissue can be produced and examined. This allows supplying the tissue with different substances in a very controlled way.
Cultivating human cells in the Petri dish is not a big challenge today. Producing artificial tissue, however, permeated by fine blood vessels, is a much more...
Faster and secure data communication: This is the goal of a new joint project involving physicists from the University of Würzburg. The German Federal Ministry of Education and Research funds the project with 14.8 million euro.
In our digital world data security and secure communication are becoming more and more important. Quantum communication is a promising approach to achieve...
On Saturday, 10 November 2018, the research icebreaker Polarstern will leave its homeport of Bremerhaven, bound for Cape Town, South Africa.
When choosing materials to make something, trade-offs need to be made between a host of properties, such as thickness, stiffness and weight. Depending on the application in question, finding just the right balance is the difference between success and failure
Now, a team of Penn Engineers has demonstrated a new material they call "nanocardboard," an ultrathin equivalent of corrugated paper cardboard. A square...
19.11.2018 | Event News
09.11.2018 | Event News
06.11.2018 | Event News
20.11.2018 | Physics and Astronomy
20.11.2018 | Medical Engineering
20.11.2018 | Physics and Astronomy