RamanNet: a lightweight convolutional neural network for bacterial identification based on Raman spectra

Bo, Zhong; Tong, Yu-Kai; Zhang, Ru; Ye, Anpei

doi:10.1039/d2ra03722j

Cited by 14 publications

(13 citation statements)

References 26 publications

(47 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ResNet architecture [50] fixes this problem by introducing skip connections. ResNet and its modifications have been successfully applied to classify Raman spectra, outperforming shallow models by a large margin, as shown by other authors [23], [51], [52], [53] and in our previous work [51].…”

Section: Introduction and Problem Statementmentioning

confidence: 63%

“…2. The ensemble architecture we propose surpasses the current state-of-the-art closed-world DNNs, which achieve accuracies of 82.2±0.3%[23], 84.7 ±0.3%[53], and 86.7± 0.4% [52], respectively. This remarkable performance motivates us to utilize an ensemble of ResNet models augmented with an SE attention mechanism not only in the closed-world setting but also in subsequent open-world settings considered in the next Sections.…”

Section: Introduction and Problem Statementmentioning

confidence: 99%

See 1 more Smart Citation

Enhancing Open-World Bacterial Raman Spectra Identification by Feature Regularization for Improved Resilience against Unknown Classes

Balytskyi,

Kalashnyk,

Hubenko

et al. 2023

Preprint

View full text Add to dashboard Cite

The combination of Deep Learning techniques and Raman spectroscopy shows great potential offering precise and prompt identification of pathogenic bacteria in clinical settings. However, the traditional closed-set classification approaches assume that all test samples belong to one of the known pathogens, and their applicability is limited since the clinical environment is inherently unpredictable and dynamic, unknown or emerging pathogens may not be included in the available catalogs. We demonstrate that the current state-of-the-art Neural Networks identifying pathogens through Raman spectra are vulnerable to unknown inputs, resulting in an uncontrollable false positive rate. To address this issue, first, we developed a novel ensemble of ResNet architectures combined with the attention mechanism which outperforms existing closed-world methods, achieving an accuracy of 87.8 ± 0.1% compared to the best available model's accuracy of 86.7 ± 0.4%. Second, through the integration of feature regularization by the Objectosphere loss function, our model achieves both high accuracy in identifying known pathogens from the catalog and effectively separates unknown samples drastically reducing the false positive rate. Finally, the proposed feature regularization method during training significantly enhances the performance of out-of-distribution detectors during the inference phase improving the reliability of the detection of unknown classes. Our novel algorithm for Raman spectroscopy enables the detection of unknown, uncatalogued, and emerging pathogens providing the flexibility to adapt to future pathogens that may emerge, and has the potential to improve the reliability of Raman-based solutions in dynamic operating environments where accuracy is critical, such as public safety applications.

show abstract

Section: Introduction and Problem Statementmentioning

confidence: 63%

Section: Introduction and Problem Statementmentioning

confidence: 99%

Enhancing Open-World Bacterial Raman Spectra Identification by Feature Regularization for Improved Resilience against Unknown Classes

Balytskyi,

Kalashnyk,

Hubenko

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…We pre-train the models on the reference data and finetune them on the finetuning data set. (c) Comparison of accuracy with different articles using the same data set. ,,, …”

Section: Methodsmentioning

confidence: 99%

“…Chi-Sing Ho and colleagues achieved an average accuracy of 82.2% in classifying 30 bacterial subtypes using ResNet-8 . Subsequently, analysts further improved the strain classification accuracy using modified ResNet models on the same open-accessed data. ,, Within these works, the Scale-Adaptive Deep Model achieved an 85.9% accuracy without data augmentation and an 86.7% accuracy with data augmentation …”

Section: Introductionmentioning

confidence: 97%

“…13 Subsequently, analysts further improved the strain classification accuracy using modified ResNet models on the same open-accessed data. 35,37,38 Within these works, the Scale-Adaptive Deep Model achieved an 85.9% accuracy without data augmentation and an 86.7% accuracy with data augmentation. 35 This work proposes a novel Combined Mutual Learning Net (CM-Net) that significantly improves Raman spectral microbial strain classification accuracy.…”

Section: ■ Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Combined Mutual Learning Net for Raman Spectral Microbial Strain Identification

Chen,

Hu,

Xue

et al. 2024

Anal. Chem.

View full text Add to dashboard Cite

Infectious diseases pose a significant threat to global health, yet traditional microbiological identification methods suffer from drawbacks, such as high costs and long processing times. Raman spectroscopy, a label-free and noninvasive technique, provides rich chemical information and has tremendous potential in fast microbial diagnoses. Here, we propose a novel Combined Mutual Learning Net that precisely identifies microbial subspecies. It demonstrated an average identification accuracy of 87.96% in an open-access data set with thirty microbial strains, representing a 5.76% improvement. 50% of the microbial subspecies accuracies were elevated by 1% to 46%, especially for E. coli 2 improved from 31% to 77%. Furthermore, it achieved a remarkable subspecies accuracy of 92.4% in the custom-built fiber-optical tweezers Raman spectroscopy system, which collects Raman spectra at a single-cell level. This advancement demonstrates the effectiveness of this method in microbial subspecies identification, offering a promising solution for microbiology diagnosis.

show abstract

Advances in Machine Learning for SERS Analysis

Froehlich,

Wouters,

Matar Abed

et al. 2024

Surface- And Tip-Enhanced Raman Scattering Spectroscopy

View full text Add to dashboard Cite

RamanNet: a lightweight convolutional neural network for bacterial identification based on Raman spectra

Abstract: We propose a novel CNN model named RamanNet for rapid and accurate identification of bacteria at the species-level based on Raman spectra. Compared to previous CNN methods, the RamanNet reached comparable results on the Bacteria-ID Raman spectral dataset.

Cited by 14 publications

References 26 publications

Enhancing Open-World Bacterial Raman Spectra Identification by Feature Regularization for Improved Resilience against Unknown Classes

Enhancing Open-World Bacterial Raman Spectra Identification by Feature Regularization for Improved Resilience against Unknown Classes

Combined Mutual Learning Net for Raman Spectral Microbial Strain Identification

Advances in Machine Learning for SERS Analysis

Contact Info

Product

Resources

About