-
A high-density polarized 3He gas-jet target for laser-plasma applications
Authors:
P. Fedorets,
C. Zheng,
R. Engels,
I. Engin,
H. Feilbach,
U. Giesen,
H. Glückler,
C. Kannis,
F. Klehr,
M. Lennartz,
H. Pfeifer,
J. Pfennings,
C. M. Schneider,
N. Schnitzler,
H. Soltner,
R. Swaczyna,
M. Büscher
Abstract:
A laser-driven spin-polarized 3He2+ beam source for nuclear-physics experiments and for the investigation of polarized nuclear fusion demands a high-density polarized 3He gas-jet target. Such a target requires a magnetic system providing a permanent homogeneous holding field for the nuclear spins plus a set of coils for adjusting the orientation of the polarization. Starting from a transport vesse…
▽ More
A laser-driven spin-polarized 3He2+ beam source for nuclear-physics experiments and for the investigation of polarized nuclear fusion demands a high-density polarized 3He gas-jet target. Such a target requires a magnetic system providing a permanent homogeneous holding field for the nuclear spins plus a set of coils for adjusting the orientation of the polarization. Starting from a transport vessel at a maximum pressure of 3 bar, the helium gas is compressed for a short time and can be injected into a laser-interaction chamber through a non-magnetic opening valve and nozzle, thus forming jets with densities of about a few 10^19 cm-3 and widths of about 1 mm. The target comprises a 3D adjustment system for precise positioning of the jet relative to the laser focus. An auxiliary gas system provides remote target operation and flushing of the gas lines with Ar gas, which helps to reduce polarization losses. The design of the target, its operation procedures and first experimental results are presented.
△ Less
Submitted 31 January, 2022;
originally announced January 2022.
-
Towards Explainable End-to-End Prostate Cancer Relapse Prediction from H&E Images Combining Self-Attention Multiple Instance Learning with a Recurrent Neural Network
Authors:
Esther Dietrich,
Patrick Fuhlert,
Anne Ernst,
Guido Sauter,
Maximilian Lennartz,
H. Siegfried Stiehl,
Marina Zimmermann,
Stefan Bonn
Abstract:
Clinical decision support for histopathology image data mainly focuses on strongly supervised annotations, which offers intuitive interpretability, but is bound by expert performance. Here, we propose an explainable cancer relapse prediction network (eCaReNet) and show that end-to-end learning without strong annotations offers state-of-the-art performance while interpretability can be included thr…
▽ More
Clinical decision support for histopathology image data mainly focuses on strongly supervised annotations, which offers intuitive interpretability, but is bound by expert performance. Here, we propose an explainable cancer relapse prediction network (eCaReNet) and show that end-to-end learning without strong annotations offers state-of-the-art performance while interpretability can be included through an attention mechanism. On the use case of prostate cancer survival prediction, using 14,479 images and only relapse times as annotations, we reach a cumulative dynamic AUC of 0.78 on a validation set, being on par with an expert pathologist (and an AUC of 0.77 on a separate test set). Our model is well-calibrated and outputs survival curves as well as a risk score and group per patient. Making use of the attention weights of a multiple instance learning layer, we show that malignant patches have a higher influence on the prediction than benign patches, thus offering an intuitive interpretation of the prediction. Our code is available at www.github.com/imsb-uke/ecarenet.
△ Less
Submitted 26 November, 2021;
originally announced November 2021.
-
Deep learning-based bias transfer for overcoming laboratory differences of microscopic images
Authors:
Ann-Katrin Thebille,
Esther Dietrich,
Martin Klaus,
Lukas Gernhold,
Maximilian Lennartz,
Christoph Kuppe,
Rafael Kramann,
Tobias B. Huber,
Guido Sauter,
Victor G. Puelles,
Marina Zimmermann,
Stefan Bonn
Abstract:
The automated analysis of medical images is currently limited by technical and biological noise and bias. The same source tissue can be represented by vastly different images if the image acquisition or processing protocols vary. For an image analysis pipeline, it is crucial to compensate such biases to avoid misinterpretations. Here, we evaluate, compare, and improve existing generative model arc…
▽ More
The automated analysis of medical images is currently limited by technical and biological noise and bias. The same source tissue can be represented by vastly different images if the image acquisition or processing protocols vary. For an image analysis pipeline, it is crucial to compensate such biases to avoid misinterpretations. Here, we evaluate, compare, and improve existing generative model architectures to overcome domain shifts for immunofluorescence (IF) and Hematoxylin and Eosin (H&E) stained microscopy images. To determine the performance of the generative models, the original and transformed images were segmented or classified by deep neural networks that were trained only on images of the target bias. In the scope of our analysis, U-Net cycleGANs trained with an additional identity and an MS-SSIM-based loss and Fixed-Point GANs trained with an additional structure loss led to the best results for the IF and H&E stained samples, respectively. Adapting the bias of the samples significantly improved the pixel-level segmentation for human kidney glomeruli and podocytes and improved the classification accuracy for human prostate biopsies by up to 14%.
△ Less
Submitted 25 May, 2021;
originally announced May 2021.
