Purpose Conventional hematoxylin and eosin (H&E) staining is destructive and largely limited to two-dimensional sections. We aimed to develop a practical virtual staining method that produces H&E-like images from label-free three-dimensional holotomography (HT) while preserving nuclear morphology and requiring only HT at inference. Methods We designed a DAPI-guided conditional diffusion model with a shared encoder and two decoder heads (H&E and DAPI). During training, the model receives HT as condition input and predicts diffusion noise for both H&E and DAPI targets using mean-squared-error objectives. DAPI is used only during training as nucleus-centric guidance. Data were acquired from the same tissue using a sequential protocol (HT imaging, then DAPI imaging, then H&E imaging). Because local nonlinear tissue deformation remains after global affine registration, HT–H&E pairs were treated as weakly paired, while HT–DAPI provided stronger local correspondence. Results Compared with CycleGAN baselines, the proposed model produced more realistic nuclear morphology and better structural consistency. On held-out test tiles, DAPI-guided diffusion achieved lower FID and KID (FID: 9.2158; KID: <InlineEquation ID="IEq1"> <EquationSource Format="TEX">$0.0091 \pm 0.0035$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mn>0.0091</mn> <mo>±</mo> <mn>0.0035</mn> </mrow> </math></EquationSource> </InlineEquation>) than CycleGAN without DAPI (FID: 14.7447; KID: <InlineEquation ID="IEq2"> <EquationSource Format="TEX">$0.0434 \pm 0.0067$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mn>0.0434</mn> <mo>±</mo> <mn>0.0067</mn> </mrow> </math></EquationSource> </InlineEquation>). Conclusions Training-only DAPI guidance improves virtual H&E generation from label-free HT without requiring DAPI during inference. This weakly paired training design reduces dependence on expensive pixel-level registration and supports scalable, nondestructive digital histopathology workflows.

High-precision label-free virtual H&E staining of 3D holotomography using DAPI-guided conditional diffusion learning

期刊论文
发表日期：

2026年5月5日

查看全文

Taeyoung Bak,
Sangwook Kim,
Daewoong Ahn,
Hyun-seok Min,
Jimin Lee

摘要

Purpose

Conventional hematoxylin and eosin (H&E) staining is destructive and largely limited to two-dimensional sections. We aimed to develop a practical virtual staining method that produces H&E-like images from label-free three-dimensional holotomography (HT) while preserving nuclear morphology and requiring only HT at inference.

Methods

We designed a DAPI-guided conditional diffusion model with a shared encoder and two decoder heads (H&E and DAPI). During training, the model receives HT as condition input and predicts diffusion noise for both H&E and DAPI targets using mean-squared-error objectives. DAPI is used only during training as nucleus-centric guidance. Data were acquired from the same tissue using a sequential protocol (HT imaging, then DAPI imaging, then H&E imaging). Because local nonlinear tissue deformation remains after global affine registration, HT–H&E pairs were treated as weakly paired, while HT–DAPI provided stronger local correspondence.

Results

Compared with CycleGAN baselines, the proposed model produced more realistic nuclear morphology and better structural consistency. On held-out test tiles, DAPI-guided diffusion achieved lower FID and KID (FID: 9.2158; KID: $0.0091 \pm 0.0035$ $0.0091 \pm 0.0035$ ) than CycleGAN without DAPI (FID: 14.7447; KID: $0.0434 \pm 0.0067$ $0.0434 \pm 0.0067$ ).

Conclusions

Training-only DAPI guidance improves virtual H&E generation from label-free HT without requiring DAPI during inference. This weakly paired training design reduces dependence on expensive pixel-level registration and supports scalable, nondestructive digital histopathology workflows.