Methods for melanoma detection
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Benefits of technology
Problems solved by technology
Method used
Image
Examples
example 1
[0080]Derivation of a 28-Marker Gene Panel
[0081]Raw probe intensities (n=6,892,960 features) from n=49 whole blood samples were used to identify genes that best discriminated between different types of melanoma samples e.g., treated versus untreated, simultaneously. A total of 28 transcripts were identified in an unbiased manner as potential markers of melanoma behavior (Table 2).
[0082]An artificial intelligence model of melanoma disease dynamics was built using normalized gene expression of these 28 markers in whole blood from Controls (n=90), Responders / Stable (n=68), and Progressive (n=66) samples. The dataset was randomly split into training (n=169) and testing (n=55) partitions for model creation and validation respectively. Five algorithms (XGB, RF, TreeBag, SVM, NNET) were identified that best predicted the training data. In the test set, each algorithm produced probability scores that predicted the sample. Each probability score reflects the “certainty” of an algorithm that ...
PUM
| Property | Measurement | Unit |
|---|---|---|
| Fraction | aaaaa | aaaaa |
| Fraction | aaaaa | aaaaa |
| Fraction | aaaaa | aaaaa |
Abstract
Description
Claims
Application Information
Login to View More 


