Mean Absolute Error Loss Research Articles

Attenuation correction and truncation completion for breast PET/MR imaging using deep learning.

Objective. Simultaneous PET/MR scanners combine the high sensitivity of MR imaging with the functional imaging of PET. However, attenuation correction of breast PET/MR imaging is technically challenging. The purpose of this study is to establish a robust attenuation correction algorithm for breast PET/MR images that relies on deep learning (DL) to recreate the missing portions of the patient's anatomy (truncation completion), as well as to provide bone information for attenuation correction from only the PET data.Approach. Data acquired from 23 female subjects with invasive breast cancer scanned with18F-fluorodeoxyglucose PET/CT and PET/MR localized to the breast region were used for this study. Three DL models, U-Net with mean absolute error loss (DLMAE) model, U-Net with mean squared error loss (DLMSE) model, and U-Net with perceptual loss (DLPerceptual) model, were trained to predict synthetic CT images (sCT) for PET attenuation correction (AC) given non-attenuation corrected (NAC) PETPET/MRimages as inputs. The DL and Dixon-based sCT reconstructed PET images were compared against those reconstructed from CT images by calculating the percent error of the standardized uptake value (SUV) and conducting Wilcoxon signed rank statistical tests.Main results. sCT images from the DLMAEmodel, the DLMSEmodel, and the DLPerceptualmodel were similar in mean absolute error (MAE), peak-signal-to-noise ratio, and normalized cross-correlation. No significant difference in SUV was found between the PET images reconstructed using the DLMSEand DLPerceptualsCTs compared to the reference CT for AC in all tissue regions. All DL methods performed better than the Dixon-based method according to SUV analysis.Significance. A 3D U-Net with MSE or perceptual loss model can be implemented into a reconstruction workflow, and the derived sCT images allow successful truncation completion and attenuation correction for breast PET/MR images.

Domain knowledge driven 3D dose prediction using moment-based loss function

Objective. To propose a novel moment-based loss function for predicting 3D dose distribution for the challenging conventional lung intensity modulated radiation therapy plans. The moment-based loss function is convex and differentiable and can easily incorporate clinical dose volume histogram (DVH) domain knowledge in any deep learning (DL) framework without computational overhead. Approach. We used a large dataset of 360 (240 for training, 50 for validation and 70 for testing) conventional lung patients with 2 Gy × 30 fractions to train the DL model using clinically treated plans at our institution. We trained a UNet like convolutional neural network architecture using computed tomography, planning target volume and organ-at-risk contours as input to infer corresponding voxel-wise 3D dose distribution. We evaluated three different loss functions: (1) the popular mean absolute error (MAE) loss, (2) the recently developed MAE + DVH loss, and (3) the proposed MAE + moments loss. The quality of the predictions was compared using different DVH metrics as well as dose-score and DVH-score, recently introduced by the AAPM knowledge-based planning grand challenge. Main results. Model with (MAE + moment) loss function outperformed the model with MAE loss by significantly improving the DVH-score (11%, p < 0.01) while having similar computational cost. It also outperformed the model trained with (MAE + DVH) by significantly improving the computational cost (48%) and the DVH-score (8%, p < 0.01). Significance. DVH metrics are widely accepted evaluation criteria in the clinic. However, incorporating them into the 3D dose prediction model is challenging due to their non-convexity and non-differentiability. Moments provide a mathematically rigorous and computationally efficient way to incorporate DVH information in any DL architecture. The code, pretrained models, docker container, and Google Colab project along with a sample dataset are available on our DoseRTX GitHub (https://github.com/nadeemlab/DoseRTX)

Domain Knowledge Driven 3D Dose Prediction Using Moment-Based Loss Function

Dose volume histogram (DVH) metrics are widely accepted evaluation criteria in the clinic. However, incorporating these metrics into deep learning dose prediction models is challenging due to their non-convexity and non-differentiability. We propose a novel moment-based loss function for predicting 3D dose distribution for the challenging conventional lung intensity modulated radiation therapy (IMRT) plans. The moment-based loss function is convex and differentiable and can easily incorporate DVH metrics in any deep learning framework without computational overhead. The moments can also be customized to reflect the clinical priorities in 3D dose prediction. For instance, using high-order moments allows better prediction in high-dose areas for serial structures. We used a large dataset of 360 (240 for training, 50 for validation and 70 for testing) conventional lung patients with 2Gy $\times$ 30 fractions to train the deep learning (DL) model using clinically treated plans at our institution. We trained a UNet like CNN architecture using computed tomography (CT), planning target volume (PTV) and organ-at-risk contours (OAR) as input to infer corresponding voxel-wise 3D dose distribution. We evaluated three different loss functions: (1) The popular Mean Absolute Error (MAE) Loss, (2) the recently developed MAE + DVH Loss, and (3) the proposed MAE + Moments Loss. The quality of the predictions was compared using different DVH metrics as well as dose-score and DVH-score, recently introduced by the AAPM knowledge-based planning grand challenge. Model with (MAE + Moment) loss function outperformed the model with MAE loss by significantly improving the DVH-score (11%, p$<$0.01) while having similar computational cost. It also outperformed the model trained with (MAE+DVH) by significantly improving the computational cost (48%) and the DVH-score (8%, p$<$0.01).

Two-step ResUp&Down generative adversarial network to reconstruct multispectral image from aerial RGB image

Convolutional neural network has brought breakthroughs on multispectral image reconstruction research. Previous work has largely focused on reconstructing MSI using the R-G-B channels from the MSI as inputs of the model. However, it’s image manipulation rather than practical use. In real application, to reconstruct multispectral image using images from RGB camera is a research that has hardly been studied. In this research, high resolution aerial RGB images are collected by drone with RGB camera and multispectral images are collected by drone with RedEdge-M multispectral Camera. Then a new two-step Generative Adversarial Network (GAN)-based reconstruction method was proposed as follows: At first, MSI and RGB images are carefully registered to make sure that pixels are one–one correspondent. Then two data sources are cropped to form dataset. After that, a novel R-MSI GAN using is proposed. It uses a ResUp&Down block to replace the ResNet block of the Generator network and it outperforms ResNet-based GAN. The experimental results show that: (1) the combination of Mean Square Error and Discriminator (MSE-D) can alleviate the problem of the high-frequency loss of generated images. (2) The root means square error (RMSE), mean relative absolute error (MRAE) and Structural Similarity (SSIM) can only reflect overall error but can’t reflect details in reconstructed images, while different bands' statistical histogram can present the total high-frequency loss of generated bands. (3) 3 indexes, which are intersection over union (IoU) based normalized difference vegetation index (NDVI)-IoU, normalized difference red edge (NDRE)-IoU and enhance vegetation index (EVI)-IoU, were defined to verify the effect of the generated MSI and they show good consistence with vegetation index map. 4 In comparisons among ResNet-based GAN, single step ResUp&Down GAN and two-step ResUp&Down GAN(T-GAN) with 3 loss functions (L1, MSE, Discriminator), the two-step ResUp&Down GAN(T-GAN) with MSE-D loss function performs best in reconstructing RGB bands. The T-GAN with L1loss-D (mean absolute error loss) performs best in reconstructing NIR and rededge bands. In summary, the proposed methods can effectively reconstruct MSI using images from RGB camera at drone based remote sensing.

A New Framework for CNN-Based Speech Enhancement in the Time Domain.

This paper proposes a new learning mechanism for a fully convolutional neural network (CNN) to address speech enhancement in the time domain. The CNN takes as input the time frames of noisy utterance and outputs the time frames of the enhanced utterance. At the training time, we add an extra operation that converts the time domain to the frequency domain. This conversion corresponds to simple matrix multiplication, and is hence differentiable implying that a frequency domain loss can be used for training in the time domain. We use mean absolute error loss between the enhanced short-time Fourier transform (STFT) magnitude and the clean STFT magnitude to train the CNN. This way, the model can exploit the domain knowledge of converting a signal to the frequency domain for analysis. Moreover, this approach avoids the well-known invalid STFT problem since the proposed CNN operates in the time domain. Experimental results demonstrate that the proposed method substantially outperforms the other methods of speech enhancement. The proposed method is easy to implement and applicable to related speech processing tasks that require time-frequency masking or spectral mapping.