Deep Learning for HCS Image Analysis

Genedata has developed an innovative high content screening (HCS) image analysis workflow based on deep learning that cuts image analysis times by an order of magnitude, while increasing data quality and reproducibility of results.

Genedata Imagence® for HCS Image Analysis:

  • Automates time consuming and repetitive tasks during image analysis set-up
  • Increases reproducibility and detects complex phenotypes by eliminating the biased selection of handcrafted features
  • Saves time by quickly being re-applied in different experimental settings
  • Seamlessly integrates with Genedata Screener for image data analysis

Watch to learn why Genedata Imagence® is an award-winning solution for HCS image analysis.

The pharmaceutical industry stands before a paradigm shift in HCS image analysis. Several recent publications have shown successful application of deep networks to the analysis of HCS images, where these revolutionary machine learning approaches outperform conventional approaches for feature extraction and phenotype classification by a wide margin. Beyond the improved quality of the results obtained, we have seen strong evidence that these approaches will enable accurate quantification of phenotypes that are difficult or impossible to measure with conventional approaches. They will also open the field to non-experts by removing barriers in the application of image analysis. Over the next 5 years, we expect Deep Learning to become the technology of choice for novel biological screening models.

Genedata partners with leading pharmaceutical companies to create the first commercially available solution for deep learning-based HCS image analysis. The new software system enables significant scale-up of HCS operations and allows decommission of existing software tools that can only be used by experts. Interested parties should contact us to find out more about the new solution and how we can collaborate. Currently we offer the following collaboration models:

1. Consulting

ディープラーニングのアプローチが既存のHCS画像解析ワークフローをどのように改善できるか確認頂けます。まず現在のHCS画像解析プロセスと典型的な実験のデータセットを確認させて下さい。データを確認後、投資する価値があるか判断可能な情報を含んだレポートを提出します。

2. Early Access

実際の実験に対しディープラーニングを使用した画像解析がどのように実施可能か確認、あるいは認識していなかった表現型を見つけます。解析に関わる問題に関係なく、コンサルティングや限定的なライセンス、あるいは両方を通じ満足頂ける答えを得るお手伝いをします。

3. License

画像解析のゴールに応じてHCSアッセイ開発やプロダクションのワークフロー、あるいはその両方にディープラーニングの適用します。弊社のソリューションへのフルアクセス、定期的な更新、メインテナンス、サポート、そして広範囲にわたるサイエンスとコンサルティングパッケージを提供します。

詳細はこちらまで。

Resources
Click to download directly or to request locked material

Application Note

Webinar

Product sheet

Product sheet

Publication (free access)

Poster

Poster

Poster

Poster

Poster

Poster