ovizio infographic machine learning 1machine learning the machine learning (ml) field uses...

1
MACHINE LEARNING The machine learning (ML) field uses artificial intelligence (AI) algorithms to allow computers to automatically “learn” how to perform a specific task without being explicitly programmed to do so. Ovizio’s services can create custom ML models to suit your specific needs. VISIT OVIZIO.COM The OsOne software includes machine learning algorithms for tracking specific objects or cell states. For example some of the current OsOne ML models allow the detection of: DEAD CELLS ACTIVATED CELLS VIRAL INFECTION BEADS CUSTOM DETECTION ... A ML model is built by showing the computer examples of data related to the desired task. 1 Data set creation 2 Model training 3 Model evaluation & prediction Type A Image acquisition & reference measurements Creation of a training set of examples to build the ML model ML model Prediction at a single-cell level Real-time measurements Cell culture parameters 81.6% R a t e o f t y p e A c e l l s 100% cell type B 100% cell type A Type A Type B Type B The effectiveness of the resulting ML model relies on the quality of the training data set. To ensure robust models, the ideal training set must be: Representative of the data to be processed Visibly different between the population types High enough quality to capture relevant cell features As balanced as possible Sufficiently large Type A Type B ?

Upload: others

Post on 30-Jun-2020

2 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Ovizio Infographic machine learning 1MACHINE LEARNING The machine learning (ML) field uses artificial intelligence (AI) algorithms to allow computers to automatically “learn”

MACHINE LEARNINGThe machine learning (ML) field uses artificial intelligence (AI) algorithms to allow computers to automatically “learn” how to perform a specific task without being explicitly programmed to do so.

Ovizio’s services can create custom ML models to suit your specific needs.

VISIT OVIZIO.COM

The OsOne software includes machine learning algorithms for tracking specific objects or cell states. For example some of the current OsOne ML models allow the detection of:

DEAD CELLS ACTIVATED CELLS VIRAL INFECTION BEADS CUSTOMDETECTION

...

A ML model is built by showing the computer examples of data related to the desired task.

1 Data set creation 2 Model training 3 Model evaluation & prediction

Type A

Image acquisition & reference measurements

Creation of a training set of examplesto build the ML model

ML model

Prediction at a single-cell level

Real-timemeasurements

Cell culture parameters

81.6%Rate

of type A cells

100% cell type B100% cell type A

Type A

Type B

Type B

The effectiveness of the resulting ML model relies on the quality of the training data set. To ensure robust models, the ideal training set must be:

Representative of the data to be processed

Visibly di�erent betweenthe population types

High enough quality to capture relevant cell features

As balanced as possible

Su�ciently large

Type A

Type B

?