Raman spectroscopy and artificial intelligence to predict the Bayesian probability of breast cancer

Improving skin cancer detection by Raman spectroscopy using convolutional neural networks and data augmentation

Background

Our previous studies have demonstrated that Raman spectroscopy could be used for skin cancer detection with good sensitivity and specificity. The objective of this study is to determine if skin cancer detection can be further improved by combining deep neural networks and Raman spectroscopy.

Patients and methods

Raman spectra of 731 skin lesions were included in this study, containing 340 cancerous and precancerous lesions (melanoma, basal cell carcinoma, squamous cell carcinoma and actinic keratosis) and 391 benign lesions (melanocytic nevus and seborrheic keratosis). One-dimensional convolutional neural networks (1D-CNN) were developed for Raman spectral classification. The stratified samples were divided randomly into training (70%), validation (10%) and test set (20%), and were repeated 56 times using parallel computing. Different data augmentation strategies were implemented for the training dataset, including added random noise, spectral shift, spectral combination and artificially synthesized Raman spectra using one-dimensional generative adversarial networks (1D-GAN). The area under the receiver operating characteristic curve (ROC AUC) was used as a measure of the diagnostic performance. Conventional machine learning approaches, including partial least squares for discriminant analysis (PLS-DA), principal component and linear discriminant analysis (PC-LDA), support vector machine (SVM), and logistic regression (LR) were evaluated for comparison with the same data splitting scheme as the 1D-CNN.

Results

The ROC AUC of the test dataset based on the original training spectra were 0.886±0.022 (1D-CNN), 0.870±0.028 (PLS-DA), 0.875±0.033 (PC-LDA), 0.864±0.027 (SVM), and 0.525±0.045 (LR), which were improved to 0.909±0.021 (1D-CNN), 0.899±0.022 (PLS-DA), 0.895±0.022 (PC-LDA), 0.901±0.020 (SVM), and 0.897±0.021 (LR) respectively after augmentation of the training dataset (p<0.0001, Wilcoxon test). Paired analyses of 1D-CNN with conventional machine learning approaches showed that 1D-CNN had a 1-3% improvement (p<0.001, Wilcoxon test).

Conclusions

Data augmentation not only improved the performance of both deep neural networks and conventional machine learning techniques by 2-4%, but also improved the performance of the models on spectra with higher noise or spectral shifting. Convolutional neural networks slightly outperformed conventional machine learning approaches for skin cancer detection by Raman spectroscopy.

Use of artificial intelligence in breast surgery: a narrative review

Background and Objective

We have witnessed tremendous advances in artificial intelligence (AI) technologies. Breast surgery, a subspecialty of general surgery, has notably benefited from AI technologies. This review aims to evaluate how AI has been integrated into breast surgery practices, to assess its effectiveness in improving surgical outcomes and operational efficiency, and to identify potential areas for future research and application.

Methods

Two authors independently conducted a comprehensive search of PubMed, Google Scholar, EMBASE, and Cochrane CENTRAL databases from January 1, 1950, to September 4, 2023, employing keywords pertinent to AI in conjunction with breast surgery or cancer. The search focused on English language publications, where relevance was determined through meticulous screening of titles, abstracts, and full-texts, followed by an additional review of references within these articles. The review covered a range of studies illustrating the applications of AI in breast surgery encompassing lesion diagnosis to postoperative follow-up. Publications focusing specifically on breast reconstruction were excluded.

Key Content and Findings

AI models have preoperative, intraoperative, and postoperative applications in the field of breast surgery. Using breast imaging scans and patient data, AI models have been designed to predict the risk of breast cancer and determine the need for breast cancer surgery. In addition, using breast imaging scans and histopathological slides, models were used for detecting, classifying, segmenting, grading, and staging breast tumors. Preoperative applications included patient education and the display of expected aesthetic outcomes. Models were also designed to provide intraoperative assistance for precise tumor resection and margin status assessment. As well, AI was used to predict postoperative complications, survival, and cancer recurrence.

Conclusions

Extra research is required to move AI models from the experimental stage to actual implementation in healthcare. With the rapid evolution of AI, further applications are expected in the coming years including direct performance of breast surgery. Breast surgeons should be updated with the advances in AI applications in breast surgery to provide the best care for their patients.

Identification of extracellular vesicles from their Raman spectra via self-supervised learning

Extracellular vesicles (EVs) released from cells attract interest for their possible role in health and diseases. The detection and characterization of EVs is challenging due to the lack of specialized methodologies. Raman spectroscopy, however, has been suggested as a novel approach for biochemical analysis of EVs. To extract information from the spectra, a novel deep learning architecture is explored as a versatile variant of autoencoders. The proposed architecture considers the frequency range separately from the intensity of the spectra. This enables the model to adapt to the frequency range, rather than requiring that all spectra be pre-processed to the same frequency range as it was trained on. It is demonstrated that the proposed architecture accepts Raman spectra of EVs and lipoproteins from 13 biological sources and from two laboratories. High reconstruction accuracy is maintained despite large variances in frequency range and noise level. It is also shown that the architecture is able to cluster the biological nanoparticles by their Raman spectra and differentiate them by their origin without pre-processing of the spectra or supervision during learning. The model performs label-free differentiation, including separating EVs from activated vs. non-activated blood platelets and EVs/lipoproteins from prostate cancer patients versus non-cancer controls. The differentiation is evaluated by creating a neural network classifier that observes the features extracted by the model to classify the spectra according to their sample origin. The classification reveals a test sensitivity of 92.2 % and selectivity of 92.3 % over 769 measurements from two labs that have different measurement configurations.

Raman-Based Techniques in Medical Applications for Diagnostic Tasks: A Review

Raman spectroscopy is a widely developing approach for noninvasive analysis that can provide information on chemical composition and molecular structure. High chemical specificity calls for developing different medical diagnostic applications based on Raman spectroscopy. This review focuses on the Raman-based techniques used in medical diagnostics and provides an overview of such techniques, possible areas of their application, and current limitations. We have reviewed recent studies proposing conventional Raman spectroscopy and surface-enhanced Raman spectroscopy for rapid measuring of specific biomarkers of such diseases as cardiovascular disease, cancer, neurogenerative disease, and coronavirus disease (COVID-19). As a result, we have discovered several most promising Raman-based applications to identify affected persons by detecting some significant spectral features. We have analyzed these approaches in terms of their potentially diagnostic power and highlighted the remaining challenges and limitations preventing their translation into clinical settings.

Computational approaches in rheumatic diseases - Deciphering complex spatio-temporal cell interactions

Inflammatory arthritis, including rheumatoid (RA), and psoriatic (PsA) arthritis, are clinically and immunologically heterogeneous diseases with no identified cure. Chronic inflammation of the synovial tissue ushers loss of function of the joint that severely impacts the patient's quality of life, eventually leading to disability and life-threatening comorbidities. The pathogenesis of synovial inflammation is the consequence of compounded immune and stromal cell interactions influenced by genetic and environmental factors. Deciphering the complexity of the synovial cellular landscape has accelerated primarily due to the utilisation of bulk and single cell RNA sequencing. Particularly the capacity to generate cell-cell interaction networks could reveal evidence of previously unappreciated processes leading to disease. However, there is currently a lack of universal nomenclature as a result of varied experimental and technological approaches that discombobulates the study of synovial inflammation. While spatial transcriptomic analysis that combines anatomical information with transcriptomic data of synovial tissue biopsies promises to provide more insights into disease pathogenesis, in vitro functional assays with single-cell resolution will be required to validate current bioinformatic applications. In order to provide a comprehensive approach and translate experimental data to clinical practice, a combination of clinical and molecular data with machine learning has the potential to enhance patient stratification and identify individuals at risk of arthritis that would benefit from early therapeutic intervention. This review aims to provide a comprehensive understanding of the effect of computational approaches in deciphering synovial inflammation pathogenesis and discuss the impact that further experimental and novel computational tools may have on therapeutic target identification and drug development.

Two-Dimensional Clustering of Spectral Changes for the Interpretation of Raman Hyperspectra

Two-dimensional correlation spectroscopy (2D-COS) is a technique that permits the examination of synchronous and asynchronous changes present in hyperspectral data. It produces two-dimensional correlation coefficient maps that represent the mutually correlated changes occurring at all Raman wavenumbers during an implemented perturbation. To focus our analysis on clusters of wavenumbers that tend to change together, we apply a k-means clustering to the wavenumber profiles in the perturbation domain decomposition of the two-dimensional correlation coefficient map. These profiles (or trends) reflect peak intensity changes as a function of the perturbation. We then plot the co-occurrences of cluster members two-dimensionally in a manner analogous to a two-dimensional correlation coefficient map. Because wavenumber profiles are clustered based on their similarity, two-dimensional cluster member spectra reveal which Raman peaks change in a similar manner, rather than how much they are correlated. Furthermore, clustering produces a discrete partitioning of the wavenumbers, thus a two-dimensional cluster member spectrum exhibits a discrete presentation of related Raman peaks as opposed to the more continuous representations in a two-dimensional correlation coefficient map. We demonstrate first the basic principles of the technique with the aid of synthetic data. We then apply it to Raman spectra obtained from a polystyrene perchlorate model system followed by Raman spectra from mammalian cells fixed with different percentages of methanol. Both data sets were designed to produce differential changes in sample components. In both cases, all the peaks pertaining to a given component should then change in a similar manner. We observed that component-based profile clustering did occur for polystyrene and perchlorate in the model system and lipids, nucleic acids, and proteins in the mammalian cell example. This confirmed that the method can translate to "real world" samples. We contrast these results with two-dimensional correlation spectroscopy results. To supplement interpretation, we present the cluster-segmented mean spectrum of the hyperspectral data. Overall, this technique is expected to be a valuable adjunct to two-dimensional correlation spectroscopy to further facilitate hyperspectral data interpretation and analysis.

Leveraging the Academic Artificial Intelligence Silecosystem to Advance the Community Oncology Enterprise

Over the last 75 years, artificial intelligence has evolved from a theoretical concept and novel paradigm describing the role that computers might play in our society to a tool with which we daily engage. In this review, we describe AI in terms of its constituent elements, the synthesis of which we refer to as the AI Silecosystem. Herein, we provide an historical perspective of the evolution of the AI Silecosystem, conceptualized and summarized as a Kuhnian paradigm. This manuscript focuses on the role that the AI Silecosystem plays in oncology and its emerging importance in the care of the community oncology patient. We observe that this important role arises out of a unique alliance between the academic oncology enterprise and community oncology practices. We provide evidence of this alliance by illustrating the practical establishment of the AI Silecosystem at the City of Hope Comprehensive Cancer Center and its team utilization by community oncology providers.

The emerging applications and advancements of Raman spectroscopy in pediatric cancers

Although the survival rate of pediatric cancer has significantly improved, it is still an important cause of death among children. New technologies have been developed to improve the diagnosis, treatment, and prognosis of pediatric cancers. Raman spectroscopy (RS) is a non-destructive analytical technique that uses different frequencies of scattering light to characterize biological specimens. It can provide information on biological components, activities, and molecular structures. This review summarizes studies on the potential of RS in pediatric cancers. Currently, studies on the application of RS in pediatric cancers mainly focus on early diagnosis, prognosis prediction, and treatment improvement. The results of these studies showed high accuracy and specificity. In addition, the combination of RS and deep learning is discussed as a future application of RS in pediatric cancer. Studies applying RS in pediatric cancer illustrated good prospects. This review collected and analyzed the potential clinical applications of RS in pediatric cancers.

Digital health: trends, opportunities and challenges in medical devices, pharma and bio-technology

Digital health interventions refer to the use of digital technology and connected devices to improve health outcomes and healthcare delivery. This includes telemedicine, electronic health records, wearable devices, mobile health applications, and other forms of digital health technology. To this end, several research and developmental activities in various fields are gaining momentum. For instance, in the medical devices sector, several smart biomedical materials and medical devices that are digitally enabled are rapidly being developed and introduced into clinical settings. In the pharma and allied sectors, digital health-focused technologies are widely being used through various stages of drug development, viz. computer-aided drug design, computational modeling for predictive toxicology, and big data analytics for clinical trial management. In the biotechnology and bioengineering fields, investigations are rapidly growing focus on digital health, such as omics biology, synthetic biology, systems biology, big data and personalized medicine. Though digital health-focused innovations are expanding the horizons of health in diverse ways, here the development in the fields of medical devices, pharmaceutical technologies and biotech sectors, with emphasis on trends, opportunities and challenges are reviewed. A perspective on the use of digital health in the Indian context is also included.

Recent Advances in Raman Spectral Imaging in Cell Diagnosis and Gene Expression Prediction

Normal and tumor regions within cancer tissue can be distinguished using various methods, such as histological analysis, tumor marker testing, X-ray imaging, or magnetic resonance imaging. Recently, new discrimination methods utilizing the Raman spectra of tissues have been developed and put into practical use. Because Raman spectral microscopy is a non-destructive and non-labeling method, it is potentially compatible for use in the operating room. In this review, we focus on the basics of Raman spectroscopy and Raman imaging in live cells and cell type discrimination, as these form the bases for current Raman scattering-based cancer diagnosis. We also review recent attempts to estimate the gene expression profile from the Raman spectrum of living cells using simple machine learning. Considering recent advances in machine learning techniques, we speculate that cancer type discrimination using Raman spectroscopy will be possible in the near future.

Classification of skin cancer using convolutional neural networks analysis of Raman spectra

BACKGROUND AND OBJECTIVE

Skin cancer is the most common malignancy in whites accounting for about one third of all cancers diagnosed per year. Portable Raman spectroscopy setups for skin cancer "optical biopsy" are utilized to detect tumors based on their spectral features caused by the comparative presence of different chemical components. However, low signal-to-noise ratio in such systems may prevent accurate tumors classification. Thus, there is a challenge to develop methods for efficient skin tumors classification.

METHODS

We compare the performance of convolutional neural networks and the projection on latent structures with discriminant analysis for discriminating skin cancer using the analysis of Raman spectra with a high autofluorescence background stimulated by a 785 nm laser. We have registered the spectra of 617 cases of skin neoplasms (615 patients, 70 melanomas, 122 basal cell carcinomas, 12 squamous cell carcinomas and 413 benign tumors) in vivo with a portable Raman setup and created classification models both for convolutional neural networks and projection on latent structures approaches. To check the classification models stability, a 10-fold cross-validation was performed for all created models. To avoid models overfitting, the data was divided into a training set (80% of spectral dataset) and a test set (20% of spectral dataset).

RESULTS

The results for different classification tasks demonstrate that the convolutional neural networks significantly (p<0.01) outperforms the projection on latent structures. For the convolutional neural networks implementation we obtained ROC AUCs of 0.96 (0.94 - 0.97; 95% CI), 0.90 (0.85-0.94; 95% CI), and 0.92 (0.87 - 0.97; 95% CI) for classifying a) malignant vs benign tumors, b) melanomas vs pigmented tumors and c) melanomas vs seborrheic keratosis respectively.

CONCLUSIONS

The performance of the convolutional neural networks classification of skin tumors based on Raman spectra analysis is higher or comparable to the accuracy provided by trained dermatologists. The increased accuracy with the convolutional neural networks implementation is due to a more precise accounting of low intensity Raman bands in the intense autofluorescence background. The achieved high performance of skin tumors classifications with convolutional neural networks analysis opens a possibility for wide implementation of Raman setups in clinical setting.

Label-free discrimination of tumorigenesis stages using in vitro prostate cancer bone metastasis model by Raman imaging

Metastatic prostate cancer colonizes the bone to pave the way for bone metastasis, leading to skeletal complications associated with poor prognosis and morbidity. This study demonstrates the feasibility of Raman imaging to differentiate between cancer cells at different stages of tumorigenesis using a nanoclay-based three-dimensional (3D) bone mimetic in vitro model that mimics prostate cancer bone metastasis. A comprehensive study comparing the classification of as received prostate cancer cells in a two-dimensional (2D) model and cancer cells in a 3D bone mimetic environment was performed over various time intervals using principal component analysis (PCA). Our results showed distinctive spectral differences in Raman imaging between prostate cancer cells and the cells cultured in 3D bone mimetic scaffolds, particularly at 1002, 1261, 1444, and 1654 cm-1, which primarily contain proteins and lipids signals. Raman maps capture sub-cellular responses with the progression of tumor cells into metastasis. Raman feature extraction via cluster analysis allows for the identification of specific cellular constituents in the images. For the first time, this work demonstrates a promising potential of Raman imaging, PCA, and cluster analysis to discriminate between cancer cells at different stages of metastatic tumorigenesis.

Raman spectroscopy: current applications in breast cancer diagnosis, challenges and future prospects

Despite significant improvements in the way breast cancer is managed and treated, it continues to persist as a leading cause of death worldwide. If detected and diagnosed early, when tumours are small and localised, there is a considerably higher chance of survival. However, current methods for detection and diagnosis lack the required sensitivity and specificity for identifying breast cancer at the asymptomatic or very early stages. Thus, there is a need to develop more rapid and reliable methods, capable of detecting disease earlier, for improved disease management and patient outcome. Raman spectroscopy is a non-destructive analytical technique that can rapidly provide highly specific information on the biochemical composition and molecular structure of samples. In cancer, it has the capacity to probe very early biochemical changes that accompany malignant transformation, even prior to the onset of morphological changes, to produce a fingerprint of disease. This review explores the application of Raman spectroscopy in breast cancer, including discussion on its capabilities in analysing both ex-vivo tissue and liquid biopsy samples, and its potential in vivo applications. The review also addresses current challenges and potential future uses of this technology in cancer research and translational clinical application.

Development of deep learning algorithms to discriminate giant cell tumors of bone from adjacent normal tissues by confocal Raman spectroscopy

Raman spectroscopy is a non-destructive analysis technique that provides detailed information about the chemical structure of tumors. Raman spectra of 52 giant cell tumors of bone (GCTB) and 21 adjacent normal tissues of formalin-fixed paraffin embedded (FFPE) and frozen specimens were obtained using a confocal Raman spectrometer and analyzed with machine learning and deep learning algorithms. We discovered characteristic Raman shifts in the GCTB specimens. They were assigned to phenylalanine and tyrosine. Based on the spectroscopic data, classification algorithms including support vector machine, k-nearest neighbors and long short-term memory (LSTM) were successfully applied to discriminate GCTB from adjacent normal tissues of both the FFPE and frozen specimens, with the accuracy ranging from 82.8% to 94.5%. Importantly, our LSTM algorithm showed the best performance in the discrimination of the frozen specimens, with a sensitivity and specificity of 93.9% and 95.1% respectively, and the AUC was 0.97. The results of our study suggest that confocal Raman spectroscopy accomplished by the LSTM network could non-destructively evaluate a tumor margin by its inherent biochemical specificity which may allow intraoperative assessment of the adequacy of tumor clearance.

Raman Spectroscopy: A Personalized Decision-Making Tool on Clinicians' Hands for In Situ Cancer Diagnosis and Surgery Guidance

Accurate in situ diagnosis and optimal surgical removal of a malignancy constitute key elements in reducing cancer-related morbidity and mortality. In surgical oncology, the accurate discrimination between healthy and cancerous tissues is critical for the postoperative care of the patient. Conventional imaging techniques have attempted to serve as adjuvant tools for in situ biopsy and surgery guidance. However, no single imaging modality has been proven sufficient in terms of specificity, sensitivity, multiplexing capacity, spatial and temporal resolution. Moreover, most techniques are unable to provide information regarding the molecular tissue composition. In this review, we highlight the potential of Raman spectroscopy as a spectroscopic technique with high detection sensitivity and spatial resolution for distinguishing healthy from malignant margins in microscopic scale and in real time. A Raman spectrum constitutes an intrinsic "molecular finger-print" of the tissue and any biochemical alteration related to inflammatory or cancerous tissue state is reflected on its Raman spectral fingerprint. Nowadays, advanced Raman systems coupled with modern instrumentation devices and machine learning methods are entering the clinical arena as adjunct tools towards personalized and optimized efficacy in surgical oncology.

Assessment of skin inflammation using near-infrared Raman spectroscopy combined with artificial intelligence analysis in an animal model

Near-infrared (NIR) Raman spectroscopy was applied to detect skin inflammation in an animal model. Artificial intelligence (AI) analysis improved prediction accuracy for skin inflammation.

Comparative Analysis of Machine Learning Algorithms on Surface Enhanced Raman Spectra of Clinical Staphylococcus Species

Raman spectroscopy (RS) is a widely used analytical technique based on the detection of molecular vibrations in a defined system, which generates Raman spectra that contain unique and highly resolved fingerprints of the system. However, the low intensity of normal Raman scattering effect greatly hinders its application. Recently, the newly emerged surface enhanced Raman spectroscopy (SERS) technique overcomes the problem by mixing metal nanoparticles such as gold and silver with samples, which greatly enhances signal intensity of Raman effects by orders of magnitudes when compared with regular RS. In clinical and research laboratories, SERS provides a great potential for fast, sensitive, label-free, and non-destructive microbial detection and identification with the assistance of appropriate machine learning (ML) algorithms. However, choosing an appropriate algorithm for a specific group of bacterial species remains challenging, because with the large volumes of data generated during SERS analysis not all algorithms could achieve a relatively high accuracy. In this study, we compared three unsupervised machine learning methods and 10 supervised machine learning methods, respectively, on 2,752 SERS spectra from 117 Staphylococcus strains belonging to nine clinically important Staphylococcus species in order to test the capacity of different machine learning methods for bacterial rapid differentiation and accurate prediction. According to the results, density-based spatial clustering of applications with noise (DBSCAN) showed the best clustering capacity (Rand index 0.9733) while convolutional neural network (CNN) topped all other supervised machine learning methods as the best model for predicting Staphylococcus species via SERS spectra (ACC 98.21%, AUC 99.93%). Taken together, this study shows that machine learning methods are capable of distinguishing closely related Staphylococcus species and therefore have great application potentials for bacterial pathogen diagnosis in clinical settings.

Review of Laser Raman Spectroscopy for Surgical Breast Cancer Detection: Stochastic Backpropagation Neural Networks

Laser Raman spectroscopy (LRS) is a highly specific biomolecular technique which has been shown to have the ability to distinguish malignant and normal breast tissue. This paper discusses significant advancements in the use of LRS in surgical breast cancer diagnosis, with an emphasis on statistical and machine learning strategies employed for precise, transparent and real-time analysis of Raman spectra. When combined with a variety of "machine learning" techniques LRS has been increasingly employed in oncogenic diagnostics. This paper proposes that the majority of these algorithms fail to provide the two most critical pieces of information required by the practicing surgeon: a probability that the classification of a tissue is correct, and, more importantly, the expected error in that probability. Stochastic backpropagation artificial neural networks inherently provide both pieces of information for each and every tissue site examined by LRS. If the networks are trained using both human experts and an unsupervised classification algorithm as gold standards, rapid progress can be made understanding what additional contextual data is needed to improve network classification performance. Our patients expect us to not simply have an opinion about their tumor, but to know how certain we are that we are correct. Stochastic networks can provide that information.