Neural Network
Also called: artificial neural network, ANN
A model made of layered interconnected units with weighted connections, adjusted during training until the network produces the desired outputs.
A neural network passes numeric inputs through layers of simple units, each applying a weight and a nonlinear function. Training adjusts millions of weights so the network's outputs match the training labels.
The biological analogy in the name is historical and loose. These units do not behave like neurons in any meaningful sense, and the resemblance has caused more confusion than insight in clinical discussions.
Where This Gets Misread
Network size is frequently quoted as a proxy for capability. It is not. A larger model trained on the wrong population performs worse on your patients than a smaller one trained on the right one, and parameter counts appear in marketing material far more often than validation details do.