Impact of deep learning image reconstruction on ADC quantification and histogram metrics: a phantom study

Key Points

• Evaluate the effects of deep learning image reconstruction on ADC quantification and histogram metrics.
• Conducted a phantom study to assess ADC accuracy across different DL levels.
• Analyzed ADC repeatability in both full and reduced field of view (FOV) diffusion-weighted imaging (DWI) sequences.
• Measured histogram metrics such as entropy and interquartile ranges at various DL strengths.
• ADC accuracy was preserved in full and reduced FOV DWI.
• High repeatability of ADC was maintained across different DL levels.
• Histogram dispersion decreased progressively with increased DL strength, especially in high-ADC vials.

Abstract