fbpx

Klinikum Stuttgart

Klinikum Stuttgart

Oct. 28, 2022
Juan Guzman
Published: Oct. 28, 2022

Researchers at the Katharinenhospital Stuttgart in Stuttgart, Germany use advanced data analysis tools to diagnose tumors. Dr. José R. Iglesias-Rozas, Associate Professor at the Universität Tübingen and researcher at the Laboratory of Neuropathology in the Institute for Pathology at Katharinenhospital, is using Palisade NeuralTools for histological classification and grading of tumors. Histological classification of tumors is based on microscopic study of the tissue. Tumor grading is a very important aspect of diagnosis since the treatment and outcome of each case depends greatly on the assigned grade.

Neural Networks Provides the Answer

Quantitative diagnostic assessments in histopathology (microscopic changes in diseased tissue) must frequently deal with uncertain information and vague linguistic terms. Final decisions are rarely based on the evaluation of a single diagnostic clue; rather, multiple pieces of evidence are routinely observed, and the certainty of combined evidence supports the final diagnosis. Neural networks analysis, which intelligently predicts outcomes based on multiple pieces of input data, is a natural fit for such medical diagnosis applications.

The Tumor Diagnosis Study

The aim of Dr. Iglesias-Rozas’s study was to predict the degree of malignancy of tumors based on ten discrete characteristics in 786 patients. Histological sections of 786 different human brain tumors were collected. Ten histological characteristics were assessed in each case, describing the presence of a specific histological feature on a scale of zero to three, with zero being the absence of the feature, and three meaning abundant presence of the feature. NeuralTools was then used to predict a malignity coefficient between 1.00 and 4.00.

"I am delighted with the program for its speed and the easy handling. I was very happy to work with numeric and category variables. The program is super quick."

Dr. José R. Iglesias-Rozas
Laboratory of Neuropathology, Institute for Pathology at Katharinenhospital

NeuralTools Predicts the Result

629 tumors were available for the NeuralTools training set, and 157 independent cases were used as the NeuralTools test set. NeuralTools accurately predicted 98.58 % of the training set cases and 95 % of the testing set cases!

According to Dr. Iglesias-Rozas, “I am delighted with the program for its speed and the easy handling. I was very happy to work with numeric and category variables.” He adds: "The program is super quick."

What’s next for NeuralTools and the study? Dr. Iglesias-Rozas explains, “We have over 30 years of data and more than 8000 patients with different brain tumors to assess next.”

Simulation Results

After running the simulations, Nagel found that at smaller plant capacities of 4 MWeq and below, the CHP plant scenarios had comparatively higher NPVs. However, larger plant capacities (5 MWeq and greater) the bio-methane plants have larger NPVs. "At a capacity of 6 MWeq, both usage pathways had their greatest potential NPV, thanks to economies of scale," Nagel explains. "Bio-methane plants are more cost-sensitive, but also more profitable at larger plant capacities."

Researchers at the Katharinenhospital Stuttgart in Stuttgart, Germany use advanced data analysis tools to diagnose tumors. Dr. José R. Iglesias-Rozas, Associate Professor at the Universität Tübingen and researcher at the Laboratory of Neuropathology in the Institute for Pathology at Katharinenhospital, is using Palisade NeuralTools for histological classification and grading of tumors. Histological classification of tumors is based on microscopic study of the tissue. Tumor grading is a very important aspect of diagnosis since the treatment and outcome of each case depends greatly on the assigned grade.

Neural Networks Provides the Answer

Quantitative diagnostic assessments in histopathology (microscopic changes in diseased tissue) must frequently deal with uncertain information and vague linguistic terms. Final decisions are rarely based on the evaluation of a single diagnostic clue; rather, multiple pieces of evidence are routinely observed, and the certainty of combined evidence supports the final diagnosis. Neural networks analysis, which intelligently predicts outcomes based on multiple pieces of input data, is a natural fit for such medical diagnosis applications.

The Tumor Diagnosis Study

The aim of Dr. Iglesias-Rozas’s study was to predict the degree of malignancy of tumors based on ten discrete characteristics in 786 patients. Histological sections of 786 different human brain tumors were collected. Ten histological characteristics were assessed in each case, describing the presence of a specific histological feature on a scale of zero to three, with zero being the absence of the feature, and three meaning abundant presence of the feature. NeuralTools was then used to predict a malignity coefficient between 1.00 and 4.00.

"I am delighted with the program for its speed and the easy handling. I was very happy to work with numeric and category variables. The program is super quick."

Dr. José R. Iglesias-Rozas
Laboratory of Neuropathology, Institute for Pathology at Katharinenhospital

NeuralTools Predicts the Result

629 tumors were available for the NeuralTools training set, and 157 independent cases were used as the NeuralTools test set. NeuralTools accurately predicted 98.58 % of the training set cases and 95 % of the testing set cases!

According to Dr. Iglesias-Rozas, “I am delighted with the program for its speed and the easy handling. I was very happy to work with numeric and category variables.” He adds: "The program is super quick."

What’s next for NeuralTools and the study? Dr. Iglesias-Rozas explains, “We have over 30 years of data and more than 8000 patients with different brain tumors to assess next.”

Simulation Results

After running the simulations, Nagel found that at smaller plant capacities of 4 MWeq and below, the CHP plant scenarios had comparatively higher NPVs. However, larger plant capacities (5 MWeq and greater) the bio-methane plants have larger NPVs. "At a capacity of 6 MWeq, both usage pathways had their greatest potential NPV, thanks to economies of scale," Nagel explains. "Bio-methane plants are more cost-sensitive, but also more profitable at larger plant capacities."

magnifierarrow-right
linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram