Computer-assisted image classification: use of neural networks in anatomic pathology

Differential counting of blood leukocytes using automated microscopy and a decision support system based on artificial neural networks--evaluation of DiffMaster Octavia

The morphological appearance of blood cells has an established association to clinical conditions. A novel system, DiffMaster Octavia for differential counting of blood leukocytes, has been evaluated. The system consisted of a microscope, 3-chip color charge coupled device (CCD) camera, automated motorized stage holder, electronic hardware for motor and light control and software for automatic cell location and image processing for preclassification of blood cells using artificial neural networks. The DiffMaster test method, was evaluated on 322 routine blood samples (400 cells per sample) using manual microscopy as reference method. The results showed a correlation of determination (r(2)) of 0.8-0.98 for the normal cell classes and blast cells. The DiffMaster correctly preclassified 89% of all leukocytes with a good reproducibility. After verification of the cell classes, the agreement between the test and reference method was 91% whether the sample was abnormal or normal. The clinical sensitivity was 98% and specificity 82%. The sensitivity to identify blast cells was slightly higher with the DiffMaster than manual microscopy. Similar levels of short-term imprecision for the two methods were found for all cell classes. In conclusion this study shows that the DiffMaster can provide a decision support system which, together with a qualified morphologist, can generate leukocyte differential count reports of high quality.

Model selection for a medical diagnostic decision support system: a breast cancer detection case

There are a number of different quantitative models that can be used in a medical diagnostic decision support system (MDSS) including parametric methods (linear discriminant analysis or logistic regression), non-parametric models (K nearest neighbor, or kernel density) and several neural network models. The complexity of the diagnostic task is thought to be one of the prime determinants of model selection. Unfortunately, there is no theory available to guide model selection. Practitioners are left to either choose a favorite model or to test a small subset using cross validation methods. This paper illustrates the use of a self-organizing map (SOM) to guide model selection for a breast cancer MDSS. The topological ordering properties of the SOM are used to define targets for an ideal accuracy level similar to a Bayes optimal level. These targets can then be used in model selection, variable reduction, parameter determination, and to assess the adequacy of the clinical measurement system. These ideas are applied to a successful model selection for a real-world breast cancer database. Diagnostic accuracy results are reported for individual models, for ensembles of neural networks, and for stacked predictors.

On the misuses of artificial neural networks for prognostic and diagnostic classification in oncology

The application of artificial neural networks (ANNs) for prognostic and diagnostic classification in clinical medicine has become very popular. In particular, feed-forward neural networks have been used extensively, often accompanied by exaggerated statements of their potential. In this paper, the essentials of feed-forward neural networks and their statistical counterparts (that is, logistic regression models) are reviewed. We point out that the uncritical use of ANNs may lead to serious problems, such as the fitting of implausible functions to describe the probability of class membership and the underestimation of misclassification probabilities. In applications of ANNs to survival data, further difficulties arise. Finally, the results of a search in the medical literature from 1991 to 1995 on applications of ANNs in oncology and some important common mistakes are reported. It is concluded that there is no evidence so far that application of ANNs represents real progress in the field of diagnosis and prognosis in oncology.