New medicine, thanks to AI
May 1, 20232006
June 1, 20231999 – 2002
Artificial neural networks replicate the biological brain system by using a set of algorithms determining the diagnostic capability of a new automated methodology...
Artificial neural networks
Artificial Neural Networks (ANNs) are neural networks that use a collection of algorithms to mimic the biological brain system. At its most basic level, a neural network consists of four main components: inputs, weights, a bias or threshold, and an output. It is possible to use it in nonlinear statistical modeling. The receiver operating characteristic technique allows neural networks to mediate predictions for individual patients with prevalence and misclassification cost considerations. The principal applications in 2001 were risk prediction for coronary heart disease, prostate cancer diagnosis, and medication dosing. (1,2)
The Second Artificial Intelligence Winter
Artificial Neural Network Analysis (ANNA) might help establish the diagnostic potential of a novel automated technique for analyzing transrectal ultrasonography (TRUS) data. This method was created to address the problem of visually distinguishing benign from malignant tissue on TRUS. The new accurate approach of computerized virtual linkage of preoperative ultrasound data and radical prostatectomy histology was designed to teach and objectively evaluate ANNA. After training with this pathologically proven digital TRUS information, ANNA was tested in blinded research. The results demonstrated that it made the diagnosis of malignant prostate tissue easier and more accurate, avoiding unnecessary operations. (3)
The Second Artificial Intelligence Winter
Artificial Neural Network Analysis (ANNA) might help establish the diagnostic potential of a novel automated technique for analyzing transrectal ultrasonography (TRUS) data. This method was created to address the problem of visually distinguishing benign from malignant tissue on TRUS.
The new accurate approach of computerized virtual linkage of preoperative ultrasound data and radical prostatectomy histology was designed to teach and objectively evaluate ANNA. After training with this pathologically proven digital TRUS information, ANNA was tested in blinded research. The results demonstrated that it made the diagnosis of malignant prostate tissue easier and more accurate, avoiding unnecessary operations. (3)