Skinscan - Low-cost 3d-scanning for dermatologic diagnosis and documentation

Merlin A Nau, Florian Schiffers, Yunhao Li, Bingjie Xu, Andreas Maier, Jack Tumblin, Marc Walton, Aggelos K Katsaggelos, Florian Willomitzer, Oliver Cossairt

ICIP 2021

my alt text
Figure 1: We propose a new architecture for holographic displays specifically designed for speckle reduction. Instead of a single coherent source of illumination, our design uses a grid of multiple sources, which sum incoherently at the image plane. By using two spatial light modulators (SLMs) with an air gap in between, we break correlations between the multiple sources enabling high-resolution holograms with significantly suppressed speckle. We experimentally demonstrate speckle reduction on both 2D images (left) and focal stacks with natural defocus blur (right).
Paper Display
Paper Thumbnail
Skinscan - Low-cost 3d-scanning for dermatologic diagnosis and documentation
Merlin A Nau, Florian Schiffers, Yunhao Li, Bingjie Xu, Andreas Maier, Jack Tumblin, Marc Walton, Aggelos K Katsaggelos, Florian Willomitzer, Oliver Cossairt
ICIP 2021
Paper

my alt text
Figure 1: Illustration of the gradient illumination procedure. Images are acquired under each gradient pattern in horizontal and vertical direction. From there the directional images are subtracted, forming x and y component of the normal map.
my alt text
Figure 3: (a) Skin section containing two small moles. b) Surface normal map of skin section. c) Banana peel with cuts and dents. d) Surface normal map the banana peel. e) 3D reconstruction of the cuts in the banana peel. f) Silicone skin replicas and their 3D reconstruction of the number "16".

Introduction

In this work, we explore the application of computational photography in the medical field, with a focus on dermatology. We propose a novel system and software framework that utilizes readily available, low-cost devices to acquire detailed 3D information of human skin using a 3D-gradient-illumination-based method. We believe this system holds substantial potential for the early-stage diagnosis and monitoring of skin diseases, particularly in densely populated or underdeveloped areas.

Method

We employ mobile gradient illumination to develop an economical measurement system for dermatological evaluation of skin. The system necessitates a programmable screen and a camera and can be implemented with a single mobile device, such as a smartphone or tablet. Additionally, we have developed an open-source software framework. This framework is object-oriented and includes base classes for essential components like the camera, projector, image reconstruction, and calibration class. It is implemented in Python and is compatible with various operating systems.

Results

We demonstrated the efficacy of our method through reconstructed surface normal maps from different objects, including human skin and a banana peel. The method was successful in capturing accurate 3D representations of the skin’s folds and wrinkles, with the results being highly repeatable and independent of environmental factors. The method was able to discern between different surface orientations and depths, even in areas with color variations, showcasing its potential in the assessment and monitoring of skin diseases.

Conclusion

In conclusion, we have introduced a method that is low-cost, mobile, and effective for 3D imaging using gradient illumination for dermatological imaging. It is versatile and user-friendly, with the open-source framework providing the necessary software programming for each component required by the method. While the primary focus has been on dermatological applications, the potential uses of this method are broad and could include applications in digital cultural heritage.

Should you have any more inquiries or need further clarification on any point, feel free to ask!

Software

SkinScan: An open-source framework for structured light 3D imaging, which includes the possibility to perform Deflectometry and Photometric Stereo measurements at home using solely commodity devices.