Training Deep Neural Networks Research Articles

Adversarial Attacks on the Interpretation of Neuron Activation Maximization

Feature visualization is one of the most popular techniques used to interpret the internal behavior of individual units of trained deep neural networks. Based on activation maximization, they consist of finding synthetic or natural inputs that maximize neuron activations. This paper introduces an optimization framework that aims to deceive feature visualization through adversarial model manipulation. It consists of finetuning a pre-trained model with a specifically introduced loss that aims to maintain model performance, while also significantly changing feature visualization. We provide evidence of the success of this manipulation on several pre-trained models for the classification task with ImageNet.

Algorithmic Foundation of Federated Learning with Sequential Data

The current analysis of federated optimization algorithms for training deep neural networks assumes that the data is non-sequential (e.g., images), which incurs a smooth loss objective. In contrast, edge devices generate lots of sequential data every day, where these sequences exhibit significant sequential correlation at different time stamps (e.g., text messages). In order to learn from such sequential data, people typically use a class of neural networks that is inherently nonsmooth, with a potentially unbounded smoothness parameter. Examples include recurrent neural networks, long-short-term memory networks, and transformers. It remains unclear how to design provably efficient algorithms for training these neural networks to learn from sequential data. My goal is to lay the algorithmic foundation of federated learning with sequential data, which contributes novel algorithms for learning from a range of real-world sequential data (e.g., natural language, electronic health record, transportation, time series, etc.) using state-of-the-art deep neural networks. In this talk, I will first motivate the problem by showing that the transformer, which is widely used for sequential data learning, has an unbounded smooth landscape. Then, I will introduce provably efficient federated deep learning algorithms in the presence of unbounded smoothness. In particular, I will introduce a few efficient algorithms for various settings of federated learning, including homogeneous data, heterogeneous data, and partial client participation. The main result is twofold. First, we show that the designed algorithms provably small computational and communication complexities. Second, we establish fundamental hardness results in the unbounded smoothness setting. Ultimately, I will discuss the future challenges of extending our research framework from small-scale neural networks to large language models.

Inconsistency-Based Data-Centric Active Open-Set Annotation

Active learning, a method to reduce labeling effort for training deep neural networks, is often limited by the assumption that all unlabeled data belong to known classes. This closed-world assumption fails in practical scenarios with unknown classes in the data, leading to active open-set annotation challenges. Existing methods struggle with this uncertainty. We introduce NEAT, a novel, computationally efficient, data-centric active learning approach for open-set data. NEAT differentiates and labels known classes from a mix of known and unknown classes, using a clusterability criterion and a consistency mea- sure that detects inconsistencies between model predictions and feature distribution. In contrast to recent learning-centric solutions, NEAT shows superior performance in active open- set annotation, as our experiments confirm. Additional details on the further evaluation metrics, implementation, and archi- tecture of our method can be found in the public document at https://arxiv.org/pdf/2401.04923.pdf.

Convolutional Channel-Wise Competitive Learning for the Forward-Forward Algorithm

The Forward-Forward (FF) Algorithm has been recently proposed to alleviate the issues of backpropagation (BP) commonly used to train deep neural networks. However, its current formulation exhibits limitations such as the generation of negative data, slower convergence, and inadequate performance on complex tasks. In this paper we take the main ideas of FF and improve them by leveraging channel-wise competitive learning in the context of convolutional neural networks for image classification tasks. A layer-wise loss function is introduced that promotes competitive learning and eliminates the need for negative data construction. To enhance both the learning of compositional features and feature space partitioning, a channel-wise feature separator and extractor block is proposed that complements the competitive learning process. Our method outperforms recent FF-based models on image classification tasks, achieving testing errors of 0.58%, 7.69%, 21.89%, and 48.77% on MNIST, Fashion-MNIST, CIFAR-10 and CIFAR-100 respectively. Our approach bridges the performance gap between FF learning and BP methods, indicating the potential of our proposed approach to learn useful representations in a layer-wise modular fashion, enabling more efficient and flexible learning. Our source code and supplementary material are available at https://github.com/andreaspapac/CwComp.

Toward More Generalized Malicious URL Detection Models

This paper reveals a data bias issue that can profoundly hinder the performance of machine learning models in malicious URL detection. We describe how such bias can be diagnosed using interpretable machine learning techniques and further argue that such biases naturally exist in the real world security data for training a classification model. To counteract these challenges, we propose a debiased training strategy that can be applied to most deep-learning based models to alleviate the negative effects of the biased features. The solution is based on the technique of adversarial training to train deep neural networks learning invariant embedding from biased data. Through extensive experimentation, we substantiate that our innovative strategy fosters superior generalization capabilities across both CNN-based and RNN-based detection models. The findings presented in this work not only expose a latent issue in the field but also provide an actionable remedy, marking a significant step forward in the pursuit of more reliable and robust malicious URL detection.

Abstract 2300: AI enabled whole exome& transcriptome liquid biopsy addressing MCED, MRD, and therapy selection on a single platform

Abstract Effective clinical management of cancer patients requires an accurate and early diagnosis, highly sensitive monitoring of minimal residual disease (MRD), and precise therapy selection. There are multiple tests available that attempt to address each of these needs independently with varying degrees of clinical utility. Caris Assure is a proprietary circulating nucleic acid sequencing platform that couples whole exome and transcriptome (WES/WTS) sequencing on white blood cells and plasma with advanced machine learning techniques to satisfy all three testing needs on one platform. This test detects SNVs, INDELs, structural variants, copy number, gene expression, tumor mutational burden (TMB), microsatellite instability (MSI), fragment length, and aneuploidy of both somatic (tumor and clonal hematopoiesis) and germline origin. Caris’ extensive database of over 350,000 tissue WES/WTS from solid malignancies was used to train deep learning neural networks to identify the molecular underpinnings of cancer. These networks were then deployed on WES/WTS data from plasma and buffy coat in pursuit of signals that can inform early detection, MRD and therapy selection. Validation studies were performed to characterize the analytic and clinical performance of Assure on over 3000 patient blood samples. These samples include ~1000 non-cancer patients (controls), ~1700 newly diagnosed patients where blood was collected at surgery (early detection), ~500 early-stage patients during adjuvant therapy at multiple time points (MRD), and ~200 locally advanced/metastatic patients where matched tissue testing was also performed (therapy selection). For early detection, stratification of blood samples from patients with stage I-IV cancer versus those with no reported cancer resulted in an AUC > 0.99 and included over 30 types of solid tumors. Notably, at 99.5% specificity, the sensitivities for stages I-IV (n= 119, 54, 50, 27) were 73%, 80%, 76%, and 89%. In the MRD setting for high-risk patients, the disease-free survival of patients whose cancers were predicted to recur was significantly shorter (39.6 mo) than those predicted not to recur (93.4 mo) (HR: 5.18, 95%CI: 2.94-9.09, p<.00001). This performance was observed across multiple lineages of cancer including but not limited to breast, colon, lung, and bladder. Lastly, for therapy selection, detection of driver mutations where blood was collected within 30 days of matched tissue demonstrated high concordance with a PPA of 93.8% and PPV of 96.8%. CHIP correction proved to be essential as ~35% percent of patients had CHIP mutations, including KRAS, BRAF, ATM, BRCA1/2, findings that could lead to improper therapy selection. Herein, we demonstrate for the first time a single liquid biopsy assay that addresses the entire continuum of care in clinical oncology with optimal diagnostic, prognostic, and predictive utility for patients and physicians. Citation Format: Jim Abraham, Valeriy Domenyuk, Maria Perdigones Borderias, Takayuki Yoshino, Elisabeth I. Heath, Emil Lou, Stephen Liu, John Marshall, Wafik S. El-Deiry, Anthony Shields, Martin Dietrich, Yoshiaki Nakamura, Takao Fujisawa, David D. Halbert, Dominic Sacchetti, Seth Stahl, Adam Stark, Sergey Klimov, Sourabh Antani, Chadi Nabhan, Jeffrey Swensen, George Poste, Matt Oberley, Milan Radovich, George W. Sledge, David Spetzler. AI enabled whole exome& transcriptome liquid biopsy addressing MCED, MRD, and therapy selection on a single platform [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2300.

A comprehensive study of a long-term creep thermo-mechanical fatigue behavior monitoring of BFRP composite pipeline using electrical capacitance sensors and deep learning algorithm

The composite pipeline is a relatively new and viable alternative pipeline to the more commonly used traditional one due to its good mechanical and fatigue properties and lower production cost. For this purpose, it is critical to assess the mechanical and fatigue performance of composite pipeline material under various working conditions, particularly for monitoring long-term creep thermo-mechanical fatigue behavior. In this paper, a long-term creep thermo-mechanical fatigue behavior in a basalt fiber reinforced polymer laminated composite pipeline is detected through an integrated expert system consisting of the electrical capacitance sensors and a deep learning algorithm. First, a multi-physics finite element model is established for the simulation of a long-term creep thermo-mechanical fatigue behavior in basalt fiber reinforced polymer composite pipelines subjected to long-term fatigue loading of internal pressure and thermal effect. Second, theoretical model results of long-term creep thermo-mechanical fatigue compliance (Sf(t)) over the time of creep are analyzed in pipeline material using the modulus degradation approach. Finally, an electrical potential change between electrical capacitance sensors electrodes corresponding to Sf(t) over the time of creep for some levels of long-term creep thermo-mechanical fatigue (Rf%) is recorded and then used in these datasets for training of the novel deep neural network based on one of the most widely used of the deep neural network families is the convolutional neural network, to predict Sf(t) in pipeline for various Rf% not included in the previous finite element model evaluation (i.e. electrical capacitance sensors technique). In this paper first is detected the long-term creep thermo-mechanical fatigue behavior for Rf%=25%,50%, and 75% from a finite element model and modulus degradation approach, and then is predicted the long-term creep thermo-mechanical fatigue behavior for Rf%=15%,35%,60%, and 85% respectively via deep neural network. The proposed method results are in good agreement with the experimental results available in the literature, thus verifying the accuracy and reliability of the proposed technique and its applicability to other different composite structures.

Adaptive Force Field Parameter Optimization for Expanding Reaction Simulations within Wide-Ranged Temperature.

Large-scale and long-term simulation of chemical reactions are key research topics in computational chemistry. However, there are still difficulties in simulating high-temperature reactions, such as polymer thermal decomposition. Herein, we introduce an adaptive potential parameter optimization framework designed to automatically fine-tune parameters, and the application of it to optimize ReaxFF parameters enhances the accuracy of chemical reaction simulations conducted at experimental temperatures. To achieve this, we leverage the power of Random Forests and interpretable machine learning techniques that enable the identification and selection of parameters that exert a substantial influence on the target attribute. By training deep neural network (NN) models, we established optimized parameter associations with reference properties. We train deep neural network (NN) models to establish the relationship between the optimized parameters and reference properties. We employ a Genetic Algorithm (GA) to utilize the surrogate NN model and the quantum mechanical targets to speed up the search for the optimal parameters. Our simulation results of resin pyrolysis show that the adaptive optimized ReaxFF can predict the peak temperature more accurately and obtain reasonable product composition under conditions that more closely resemble experimental scenarios. This work facilitates advances in force field parameter optimization for more accurate and universal reaction simulations.

An optimized neural network with AdaHessian for cryptojacking attack prediction for Securing Crypto Exchange Operations of MEC applications

Bitcoin exchange security is crucial because of MEC's widespread use. Cryptojacking has compromised MEC app security and bitcoin exchange ecosystem functionality. This paper propose a cutting-edge neural network and AdaHessian optimization technique for cryptojacking prediction and defense. We provide a cutting-edge deep neural network (DNN) cryptojacking attack prediction approach employing pruning, post-training quantization, and AdaHessian optimization. To solve these problems, this paper apply pruning, post-training quantization, and AdaHessian optimization. A new framework for quick DNN training utilizing AdaHessian optimization can detect cryptojacking attempts with reduced computational cost. Pruning and post-training quantization improve the model for low-CPU on-edge devices. The proposed approach drastically decreases model parameters without affecting Cryptojacking attack prediction. The model has Recall 98.72%, Precision 98.91%, F1-Score 99.09%, MSE 0.0140, RMSE 0.0137, and MAE 0.0139. Our solution beats state-of-the-art approaches in precision, computational efficiency, and resource consumption, allowing more realistic, trustworthy, and cost-effective machine learning models. We address increasing cybersecurity issues holistically by completing the DNN optimization-security loop. Securing Crypto Exchange Operations delivers scalable and efficient Cryptojacking protection, improving machine learning, cybersecurity, and network management.

Reusing Convolutional Neural Network Models through Modularization and Composition

With the widespread success of deep learning technologies, many trained deep neural network (DNN) models are now publicly available. However, directly reusing the public DNN models for new tasks often fails due to mismatching functionality or performance. Inspired by the notion of modularization and composition in software reuse, we investigate the possibility of improving the reusability of DNN models in a more fine-grained manner. Specifically, we propose two modularization approaches named CNNSplitter and GradSplitter, which can decompose a trained convolutional neural network (CNN) model for N -class classification into N small reusable modules. Each module recognizes one of the N classes and contains a part of the convolution kernels of the trained CNN model. Then, the resulting modules can be reused to patch existing CNN models or build new CNN models through composition. The main difference between CNNSplitter and GradSplitter lies in their search methods: the former relies on a genetic algorithm to explore search space, while the latter utilizes a gradient-based search method. Our experiments with three representative CNNs on three widely used public datasets demonstrate the effectiveness of the proposed approaches. Compared with CNNSplitter, GradSplitter incurs less accuracy loss, produces much smaller modules (19.88% fewer kernels), and achieves better results on patching weak models. In particular, experiments on GradSplitter show that (1) by patching weak models, the average improvement in terms of precision, recall, and F1-score is 17.13%, 4.95%, and 11.47%, respectively, and (2) for a new task, compared with the models trained from scratch, reusing modules achieves similar accuracy (the average loss of accuracy is only 2.46%) without a costly training process. Our approaches provide a viable solution to the rapid development and improvement of CNN models.

Temporal teacher with masked transformers for semi-supervised action proposal generation

By conditioning on unit-level predictions, anchor-free models for action proposal generation have displayed impressive capabilities, such as having a lightweight architecture. However, task performance depends significantly on the quality of data used in training, and most effective models have relied on human-annotated data. Semi-supervised learning, i.e., jointly training deep neural networks with a labeled dataset as well as an unlabeled dataset, has made significant progress recently. Existing works have either primarily focused on classification tasks, which may require less annotation effort, or considered anchor-based detection models. Inspired by recent advances in semi-supervised methods on anchor-free object detectors, we propose a teacher-student framework for a two-stage action detection pipeline, named Temporal Teacher with Masked Transformers (TTMT), to generate high-quality action proposals based on an anchor-free transformer model. Leveraging consistency learning as one self-training technique, the model jointly trains an anchor-free student model and a gradually progressing teacher counterpart in a mutually beneficial manner. As the core model, we design a Transformer-based anchor-free model to improve effectiveness for temporal evaluation. We integrate bi-directional masks and devise encoder-only Masked Transformers for sequences. Jointly training on boundary locations and various local snippet-based features, our model predicts via the proposed scoring function for generating proposal candidates. Experiments on the THUMOS14 and ActivityNet-1.3 benchmarks demonstrate the effectiveness of our model for temporal proposal generation task.

Implementation of singularity-free inverse kinematics for humanoid robotic arm using Bayesian optimized deep neural network

This study presents the efficacy of deep learning techniques in controlling the arm of a humanoid robot without resorting to inverse kinematic analysis. Emphasizing real-time applicability, a straightforward deep neural network (DNN) structure is developed and optimized using hyperparameter Bayesian optimization. The Keras Bayesian Optimization Tuner with Gaussian process fine-tunes the DNN architecture. Utilizing a simulation environment and generating around 10 billion datasets, the optimized DNN is evaluated on both simulated and real robots. Impressively, the trained DNN exhibits a notable ability to predict robot control within a 25 ms timeframe, achieving a total Mean Absolute Error (MAE) of 0.02 for the xyz-axes. These results underscore the potential of deep learning-based approaches in humanoid robot arm control, eliminating the need for inverse kinematic equations and providing valuable insights for future robotic system development.

Brain tumor classification for MRI images using dual-discriminator conditional generative adversarial network

ABSTRACT This research focuses on improving the detection and classification of brain tumors using a method called Brain Tumor Classification using Dual-Discriminator Conditional Generative Adversarial Network (DDCGAN) for MRI images. The proposed system is implemented in the MATLAB programming language. In this study, images of the brain are taken from a dataset and processed to remove noise and enhance image quality. The brain pictures are taken from Brats MRI image dataset. The images are preprocessed using Structural interval gradient filtering to remove noises and improve the quality of the image. The preprocessing outcomes are given to feature extraction. The features are extracted by Empirical wavelet transform (EWT) and the extracted features are given to the Dual-discriminator conditional generative adversarial network (DDCGAN) for recognizing the brain tumor, which classifies the brain images into glioma, meningioma, pituitary gland, and normal. Then, the weight parameter of DDCGAN is optimized by utilizing Border Collie Optimization (BCO), which is a met a heuristic approach to handle the real world optimization issues. It maximizes the detection accurateness and reduced computational time. Implemented in MATLAB, the experimental results demonstrate that the proposed system achieves a high sensitivity of 99.58%. The BCO-DDCGAN-MRI-BTC method outperforms existing techniques in terms of precision and sensitivity when compared to methods like Kernel Basis SVM (KSVM-HHO-BTC), Joint Training of Two-Channel Deep Neural Network (JT-TCDNN-BTC), and YOLOv2 including Convolutional Neural Network (YOLOv2-CNN-BTC). The research findings indicate that the proposed method enhances the accuracy of brain tumor classification while reducing computational time and errors.

A novel hybrid machine learning model for rapid assessment of wave and storm surge responses over an extended coastal region

Storm surge and waves are responsible for a substantial portion of tropical and extratropical cyclones-related damages. While high-fidelity numerical models have significantly advanced the simulation accuracy of storm surge and waves, they are not practical to be employed for probabilistic analysis, risk assessment or rapid prediction due to their high computational demands. In this study, a novel hybrid model combining dimensionality reduction and data-driven techniques is developed for rapid assessment of waves and storm surge responses over an extended coastal region. Specifically, the hybrid model simultaneously identifies a low-dimensional representation of the high-dimensional spatial system based on a deep autoencoder (DAE) while mapping the storm parameters to the obtained low-dimensional latent space using a deep neural network (DNN). To train the hybrid model, a combined weighted loss function is designed to encourage a balance between DAE and DNN training and achieve the best accuracy. The performance of the hybrid model is evaluated through a case study using the synthetic data from the North Atlantic Comprehensive Coastal Study (NACCS) covering critical regions within New York and New Jersey. In addition, the proposed approach is compared with two decoupled models where the regression model is based on DNN and the reduction techniques are either principal component analysis (PCA) or DAE which are trained separately from the DNN model. High accuracy and computational efficiency are observed for the hybrid model which could be readily implemented as part of early warning systems or probabilistic risk assessment of waves and storm surge.

An Efficient Checkpoint Strategy for Federated Learning on Heterogeneous Fault-Prone Nodes

Federated learning (FL) is a distributed machine learning method in which client nodes train deep neural network models locally using their own training data and then send that trained model to a server, which then aggregates all of the trained models into a globally trained model. This protects personal information while enabling machine learning with vast amounts of data through parallel learning. Nodes that train local models are typically mobile or edge devices from which data can be easily obtained. These devices typically run on batteries and use wireless communication, which limits their power, making their computing performance and reliability significantly lower than that of high-performance computing servers. Therefore, training takes a long time, and if something goes wrong, the client may have to start training again from the beginning. If this happens frequently, the training of the global model may slow down and the final performance may deteriorate. In a general computing system, a checkpointing method can be used to solve this problem, but applying an existing checkpointing method to FL may result in excessive overheads. This paper proposes a new FL method for situations with many fault-prone nodes that efficiently utilizes checkpoints.

High resolution dynamic ultrasound atlas of embryonic and fetal development of the common marmoset.

The common marmoset (Callithrix jacchus) provides an ideal model to study early development of primates, and an in vivo platform to validate conclusions from in vitro studies of human embryos and embryo models. Currently, however,no established staging atlas of marmoset embryonic developmentexists. Using high-resolution, longitudinal ultrasound scans on live pregnant marmosets, we present the first dynamic in vivo imaging of entire primate gestation beginning with attachment until the last day before birth. Our study unveils the first dynamic images of an in vivo attached mammalian embryo developing in utero, and the intricacies of the delayed development period unique to the common marmoset amongst primates, revealing a window for somatic interventions. Established obstetric and embryologic measurements for each scan were used comparatively with the standardized Carnegie staging of human development to highlight similarities and differences. Our study also allows for tracking the development of major organs. We focus on the ontogeny of the primate heart and brain. Finally, input ultrasound images were used to train deep neural networks to accurately determine the gestational age. All our ultrasounds and staging data recording are posted online so that the atlas can be used as a community resource toward monitoring and managing marmoset breeding colonies. The temporal and spatial resolution of ultrasound achieved in this study demonstrates the promise of noninvasive imaging in the marmoset for the in vivo study of primate-specific aspects of embryonic and fetal development.

Optimal Re-Materialization Strategies for Heterogeneous Chains: How to Train Deep Neural Networks with Limited Memory

Training in Feed Forward Deep Neural Networks is a memory-intensive operation which is usually performed on GPUs with limited memory capacities. This may force data scientists to limit the depth of the models or the resolution of the input data if data does not fit in the GPU memory. The re-materialization technique, whose idea comes from the checkpointing strategies developed in the Automatic Differentiation literature, allows data scientists to limit the memory requirements related to the storage of intermediate data (activations), at the cost of an increase in the computational cost. This paper introduces a new strategy of re-materialization of activations that significantly reduces memory usage. It consists in selecting which activations are saved and which activations are deleted during the forward phase, and then recomputing the deleted activations when they are needed during the backward phase. We propose an original computation model that combines two types of activation savings: either only storing the layer inputs, or recording the complete history of operations that produced the outputs. This paper focuses on the fully heterogeneous case, where the computation time and the memory requirement of each layer is different. We prove that finding the optimal solution is NP-hard and that classical techniques from Automatic Differentiation literature do not apply. Moreover, the classical assumption of memory persistence of materialized activations, used to simplify the search of optimal solutions, does not hold anymore. Thus, we propose a weak memory persistence property and provide a Dynamic Program to compute the optimal sequence of computations. This algorithm is made available through the Rotor software, a PyTorch plug-in dealing with any network consisting of a sequence of layers, each of them having an arbitrarily complex structure. Through extensive experiments, we show that our implementation consistently outperforms existing re-materialization approaches for a large class of networks, image sizes and batch sizes.

Bio-Inspired Dark Adaptive Nighttime Object Detection.

Nighttime object detection is challenging due to dim, uneven lighting. The IIHS research conducted in 2022 shows that pedestrian anti-collision systems are less effective at night. Common solutions utilize costly sensors, such as thermal imaging and LiDAR, aiming for highly accurate detection. Conversely, this study employs a low-cost 2D image approach to address the problem by drawing inspiration from biological dark adaptation mechanisms, simulating functions like pupils and photoreceptor cells. Instead of relying on extensive machine learning with day-to-night image conversions, it focuses on image fusion and gamma correction to train deep neural networks for dark adaptation. This research also involves creating a simulated environment ranging from 0 lux to high brightness, testing the limits of object detection, and offering a high dynamic range testing method. Results indicate that the dark adaptation model developed in this study improves the mean average precision (mAP) by 1.5-6% compared to traditional models. Our model is capable of functioning in both twilight and night, showcasing academic novelty. Future developments could include using virtual light in specific image areas or integrating with smart car lighting to enhance detection accuracy, thereby improving safety for pedestrians and drivers.

Classifying electrocardiograph waveforms using trained deep learning neural network based on wavelet representation

<span lang="EN-US">Due to the rise in cardiac patients, an automated system that can identify different heart disorders has been created to lighten and distribute the duty of physicians. This research uses three different electrocardiograph (ECG) signals as indicators of a person's cardiac problems: Normal sinus rhythm (NSR), arrhythmia (ARR), and congestive heart failure (CHF). The continuous wavelet transform (CWT) provides the mechanism for classifying the 190 individual cases of ECG data into a 2-dimensional time-frequency representation. In this paper, the modified GoogLeNet is used for ECG data classification. Using a transfer learning approach and adjustments to parts of the output layers, ECG classification was conducted and the effectiveness of convolutional neural network (CNN) designs was tested. By comparing the results that the optimized neural network and GoogLeNet both had classification accuracy about of 80% and 100%, respectively. The GoogLeNet provide the best result in term of accuracy and training time.</span>

Combination of DCE-MRI and NME-DWI via Deep Neural Network for Predicting Breast Cancer Molecular Subtypes

BackgroundTo explore whether the combination of dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) and nonmono-exponential (NME) model-based diffusion-weighted imaging (DWI) via deep neural network (DNN) can improve the prediction of breast cancer molecular subtypes compared to either imaging technique used alone. Patients and MethodsThis prospective study examined 480 breast cancers in 475 patients undergoing DCE-MRI and NME-DWI at 3.0 T. Breast cancers were classified as follows: human epidermal growth factor receptor 2 enriched (HER2-enriched), luminal A, luminal B (HER2–), luminal B (HER2+), and triple-negative subtypes. A total of 20% cases were withheld as an independent test dataset, and the remaining cases were used to train DNN with an 80% to 20% training-validation split and 5-fold cross-validation. The diagnostic accuracies of DNN in 5-way subtype classification between the DCE-MRI, NME-DWI, and their combined multiparametric-MRI datasets were compared using analysis of variance with least significant difference posthoc test. Areas under the receiver-operating characteristic curves were calculated to assess the performances of DNN in binary subtype classification between the 3 datasets. ResultsThe 5-way classification accuracies of DNN on both DCE-MRI (0.71) and NME-DWI (0.64) were significantly lower (P < .05) than on multiparametric-MRI (0.76), while on DCE-MRI was significantly higher (P < .05) than on NME-DWI. The comparative results of binary classification between the 3 datasets were consistent with the 5-way classification. ConclusionThe combination of DCE-MRI and NME-DWI via DNN achieved a significant improvement in breast cancer molecular subtype prediction compared to either imaging technique used alone. Additionally, DCE-MRI outperformed NME-DWI in differentiating subtypes.