Highly accurate diagnosis of lung adenocarcinoma and squamous cell carcinoma tissues by deep learning

The diagnosis value of dual-energy computed tomography (DECT) multi-parameter imaging in lung adenocarcinoma and squamous cell carcinoma

BACKGROUND

Lung cancer continues to pose a serious risk to human health. With a high mortality rate, non-small cell lung cancer (NSCLC) is the major type of lung cancer, making up to 85% of all cases of lung cancer. Lung adenocarcinoma (AC), and lung squamous cell carcinoma (SC) are the two primary types of NSCLC. Determining the pathological type of NSCLC is important in establishing the most effective treatment method. Dual-energy computed tomography (DECT) multi-parameter imaging is an imaging technology that provides accurate and reliable disease diagnosis, and its uses are utilized for the combined diagnostic efficacy of AC and SC. The purpose of this study was to investigate the diagnostic value of spectral parameters of DECT in efficacy to AC and SC, and their combined diagnostic efficacy was also analyzed.

METHODS

We conducted a retrospective analysis of clinical and imaging data for 36 patients diagnosed with SC and 35 patients with AC. These patients underwent preoperative DECT chest scans, encompassing both arterial and venous phases, at our hospital from December 2020 to April 2022. The tumor diameter, water concentration (WC), iodine concentration (IC), normalized iodine concentration (NIC), Z effective (Zeff), and slope of the curve (K) in lesions were evaluated during two scanning phases in the two separate pathological types of lung cancers. The differences in parameters between these two types of lung cancers were statistically analyzed. In addition, receiver operating characteristic (ROC) curves were performed for these parameters to distinguish between SC and AC.

RESULTS

In a univariate analysis involving 71 lung cancer patients, the results from Zeff, IC, NIC, and K from the AC's arterial and venous phase images were more elevated than those from the SC (P < 0.05). In contrast, the WC results were lower than those from SC (P < 0.05). The area under the ROC curve (AUC) for multi-parameter joint prediction typing was 0.831, with a corresponding sensitivity of 63.9% and specificity of 94.3%.

CONCLUSION

It is possible to distinguish between central SC and AC using the spectrum characteristics of DECT-enhanced scanning (Zeff, IC, NIC, K, WC, and tumor diameter). Diagnostic effectiveness can be greatly improved when multiple variables are included.

Differentiation of pathological subtypes and Ki-67 and TTF-1 expression by dual-energy CT (DECT) volumetric quantitative analysis in non-small cell lung cancer

BACKGROUND

To explore the value of dual-energy computed tomography (DECT) in differentiating pathological subtypes and the expression of immunohistochemical markers Ki-67 and thyroid transcription factor 1 (TTF-1) in patients with non-small cell lung cancer (NSCLC).

METHODS

Between July 2022 and May 2024, patients suspected of lung cancer who underwent two-phase contrast-enhanced DECT were prospectively recruited. Whole-tumor volumetric and conventional spectral analysis were utilized to measure DECT parameters in the arterial and venous phase. The DECT parameters model, clinical-CT radiological features model, and combined prediction model were developed to discriminate pathological subtypes and predict Ki-67 or TTF-1 expression. Multivariate logistic regression analysis was used to identify independent predictors. The diagnostic efficacy was assessed by the area under the receiver operating characteristic curve (AUC) and compared using DeLong's test.

RESULTS

This study included 119 patients (92 males and 27 females; mean age, 63.0 ± 9.4 years) who was diagnosed with NSCLC. When applying the DECT parameters model to differentiate between adenocarcinoma and squamous cell carcinoma, ROC curve analysis indicated superior diagnostic performance for conventional spectral analysis over volumetric spectral analysis (AUC, 0.801 vs. 0.709). Volumetric spectral analysis exhibited higher diagnostic efficacy in predicting immunohistochemical markers compared to conventional spectral analysis (both P < 0.05). For Ki-67 and TTF-1 expression, the combined prediction model demonstrated optimal diagnostic performance with AUC of 0.943 and 0.967, respectively.

CONCLUSIONS

The combined predictive model based on volumetric quantitative analysis in DECT offers valuable information to discriminate immunohistochemical expression status, facilitating clinical decision-making for patients with NSCLC.

Label-Free Surface-Enhanced Raman Spectroscopy with Machine Learning for the Diagnosis of Thyroid Cancer by Using Fine-Needle Aspiration Liquid Samples

The incidence of thyroid cancer is increasing worldwide. Fine-needle aspiration (FNA) cytology is widely applied with the use of extracted biological cell samples, but current FNA cytology is labor-intensive, time-consuming, and can lead to the risk of false-negative results. Surface-enhanced Raman spectroscopy (SERS) combined with machine learning algorithms holds promise for cancer diagnosis. In this study, we develop a label-free SERS liquid biopsy method with machine learning for the rapid and accurate diagnosis of thyroid cancer by using thyroid FNA washout fluids. These liquid supernatants are mixed with silver nanoparticle colloids, and dispersed in quartz capillary for SERS measurements to discriminate between healthy and malignant samples. We collect Raman spectra of 36 thyroid FNA samples (18 malignant and 18 benign) and compare four classification models: Principal Component Analysis-Linear Discriminant Analysis (PCA-LDA), Random Forest (RF), Support Vector Machine (SVM), and Convolutional Neural Network (CNN). The results show that the CNN algorithm is the most precise, with a high accuracy of 88.1%, sensitivity of 87.8%, and the area under the receiver operating characteristic curve of 0.953. Our approach is simple, convenient, and cost-effective. This study indicates that label-free SERS liquid biopsy assisted by deep learning models holds great promise for the early detection and screening of thyroid cancer.

The application of Raman spectroscopy for the diagnosis and monitoring of lung tumors

Raman spectroscopy is an optical technique that uses inelastic light scattering in response to vibrating molecules to produce chemical fingerprints of tissues, cells, and biofluids. Raman spectroscopy strategies produce high levels of chemical specificity without requiring extensive sample preparation, allowing for the use of advanced optical tools such as microscopes, fiber optics, and lasers that operate in the visible and near-infrared spectral range, making them increasingly suitable for a wide range of medical diagnostic applications. Metal nanoparticles and nonlinear optical effects can improve Raman signals, and optimized fiber optic Raman probes can make real-time, in vivo, single-point observations. Furthermore, diagnostic speed and spatial accuracy can be improved through the multimodal integration of Raman measurements and other technologies. Recent studies have significantly contributed to the improvement of diagnostic speed and accuracy, making them suitable for clinical application. Lung cancer is a prevalent type of respiratory malignancy. However, the use of computed tomography for detection and screening frequently reveals numerous smaller lung nodules, which makes the diagnostic process more challenging from a clinical perspective. While the majority of small nodules detected are benign, there are currently no direct methods for identifying which nodules represent very early-stage lung cancer. Positron emission tomography and other auxiliary diagnostic methods for non-surgical biopsy samples from these small nodules yield low detection rates, which might result in significant expenses and the possibility of complications for patients. While certain subsets of patients can undergo curative treatment, other individuals have a less favorable prognosis and need alternative therapeutic interventions. With the emergence of new methods for treating cancer, such as immunotherapies, which can potentially extend patient survival and even lead to a complete cure in certain instances, it is crucial to determine the most suitable biomarkers and metrics for assessing the effectiveness of these novel compounds. This will ensure that significant treatment outcomes are accurately measured. This review provides a comprehensive overview of the prospects of Raman spectroscopy and its applications in the diagnosis and analysis of lung tumors.

RaT: Raman Transformer for highly accurate melanoma detection with critical features visualization

Melanoma is an important cause of death from skin cancer. Early and accurate diagnosis can effectively reduce mortality. But the current diagnosis relies on the experience of pathologists, increasing the rate of misdiagnosis. In this paper, Raman Transformer (RaT) model is proposed by combining Raman spectroscopy and a Transformer encoder to distinguish the Raman spectra of melanoma and normal tissue. To make the spectral data more suitable for the Transformer encoder, we split the Raman spectrum into segments and map them into block vectors, which are then input into the Transformer encoder and classified using the multi-head self-attention mechanism and the Multilayer Perceptron (MLP). The RaT model achieves 99.69% accuracy, 99.61% sensitivity, and 99.82% specificity, which is higher than the classical principal component analysis with the neural network (PCA + NNET) method. In addition, we visualize and explain the fingerprint peaks found by the RaT model and their corresponding biological information. Our proposed RaT model provides a novel and reliable method for processing Raman spectral data, which is expected to help distinguish melanoma from normal cells, diagnose other diseases, and save human lives.

Label-free SERS detection of prostate cancer based on multi-layer perceptron surrogate model method

Prior surface-enhanced Raman spectroscopy (SERS) research has shown that pre-processing is necessary before analysis. Pre-processing also typically serves the dual purposes of removing the auto-fluorescence background and minimizing data volatility. This method allows for a more accurate comparison of spectral traits and relative SERS peak strength. However, because there are so many different kinds of samples, it can take a long time, and there is no assurance that the approach chosen will work well with a particular kind of sample. Therefore, this study employed a deep learning technique called multi-layer perceptron (MLP) to simplify the pre-processing of blood plasma SERS samples in patients with prostate cancer (PC), as well as to enhance the sensitivity and specificity of diagnosis using SERS technology. First of all, significant variations in peak intensity can be observed in the difference spectra, facilitating differentiation between PC and normal groups. Second, the data analysis was carried out in three different stages (raw data, defluorescenced data, and normalized data) using principal component analysis and linear discriminant analysis (PCA-LDA), as well as PCA-multi-layer perceptron (PCA-MLP). Finally, when SERS data was analyzed using PCA-LDA, there were significant differences in classification accuracy across each stage (The classification accuracy of three different stages were 76.90%, 85.60%, 95.20%, respectively). However, when PCA-MLP was utilized for SERS data analysis, the classification accuracy remained consistently high and stable (The classification accuracy of three different stages were 92.00%, 92.40%, 96.70%, respectively). The experimental results of PCA-MLP for classifying specific SERS data indicate that analyzing raw data directly can simplify the experimental process and enhance the efficacy of SERS analysis.

An eXplainable Artificial Intelligence analysis of Raman spectra for thyroid cancer diagnosis

Raman spectroscopy shows great potential as a diagnostic tool for thyroid cancer due to its ability to detect biochemical changes during cancer development. This technique is particularly valuable because it is non-invasive and label/dye-free. Compared to molecular tests, Raman spectroscopy analyses can more effectively discriminate malignant features, thus reducing unnecessary surgeries. However, one major hurdle to using Raman spectroscopy as a diagnostic tool is the identification of significant patterns and peaks. In this study, we propose a Machine Learning procedure to discriminate healthy/benign versus malignant nodules that produces interpretable results. We collect Raman spectra obtained from histological samples, select a set of peaks with a data-driven and label independent approach and train the algorithms with the relative prominence of the peaks in the selected set. The performance of the considered models, quantified by area under the Receiver Operating Characteristic curve, exceeds 0.9. To enhance the interpretability of the results, we employ eXplainable Artificial Intelligence and compute the contribution of each feature to the prediction of each sample.

Evaluation and Exploration of Machine Learning and Convolutional Neural Network Classifiers in Detection of Lung Cancer from Microarray Gene-A Paradigm Shift

Microarray gene expression-based detection and classification of medical conditions have been prominent in research studies over the past few decades. However, extracting relevant data from the high-volume microarray gene expression with inherent nonlinearity and inseparable noise components raises significant challenges during data classification and disease detection. The dataset used for the research is the Lung Harvard 2 Dataset (LH2) which consists of 150 Adenocarcinoma subjects and 31 Mesothelioma subjects. The paper proposes a two-level strategy involving feature extraction and selection methods before the classification step. The feature extraction step utilizes Short Term Fourier Transform (STFT), and the feature selection step employs Particle Swarm Optimization (PSO) and Harmonic Search (HS) metaheuristic methods. The classifiers employed are Nonlinear Regression, Gaussian Mixture Model, Softmax Discriminant, Naive Bayes, SVM (Linear), SVM (Polynomial), and SVM (RBF). The two-level extracted relevant features are compared with raw data classification results, including Convolutional Neural Network (CNN) methodology. Among the methods, STFT with PSO feature selection and SVM (RBF) classifier produced the highest accuracy of 94.47%.

Artificial Intelligence-Aided Optical Imaging for Cancer Theranostics

The use of artificial intelligence (AI) to assist biomedical imaging have demonstrated its high accuracy and high efficiency in medical decision-making for individualized cancer medicine. In particular, optical imaging methods are able to visualize both the structural and functional information of tumors tissues with high contrast, low cost, and noninvasive property. However, no systematic work has been performed to inspect the recent advances on AI-aided optical imaging for cancer theranostics. In this review, we demonstrated how AI can guide optical imaging methods to improve the accuracy on tumor detection, automated analysis and prediction of its histopathological section, its monitoring during treatment, and its prognosis by using computer vision, deep learning and natural language processing. By contrast, the optical imaging techniques involved mainly consisted of various tomography and microscopy imaging methods such as optical endoscopy imaging, optical coherence tomography, photoacoustic imaging, diffuse optical tomography, optical microscopy imaging, Raman imaging, and fluorescent imaging. Meanwhile, existing problems, possible challenges and future prospects for AI-aided optical imaging protocol for cancer theranostics were also discussed. It is expected that the present work can open a new avenue for precision oncology by using AI and optical imaging tools.

A one-dimensional convolutional neural network based deep learning for high accuracy classification of transformation stages in esophageal squamous cell carcinoma tissue using micro-FTIR

Among the most frequently diagnosed cancers in developing countries, esophageal squamous cell carcinoma (ESCC) ranks among the top six causes of death. It would be beneficial if a rapid, accurate, and automatic ESCC diagnostic method could be developed to reduce the workload of pathologists and improve the effectiveness of cancer treatments. Using micro-FTIR spectroscopy, this study classified the transformation stages of ESCC tissues. Based on 6,352 raw micro-FTIR spectra, a one-dimensional convolutional neural network (1D-CNN) model was constructed to classify-five stages. Based on the established model, more than 93% accuracy was achieved at each stage, and the accuracy of identifying proliferation, low grade neoplasia, and ESCC cancer groups was achieved 99% for the test dataset. In this proof-of-concept study, the developed method can be applied to other diseases in order to promote the use of FTIR spectroscopy in cancer pathology.

Rapid proximate analysis of coal based on reflectance spectroscopy and deep learning

Proximate analysis of coal is of profound significance for understanding coal quality and promoting rational utilization of coal resources. Traditional coal proximate analysis mainly uses chemical analysis methods, which have the disadvantages of slow speed and high cost. This paper proposed an approach combining reflectance spectroscopy with deep learning (DL) for rapid proximate analysis of coal. First, 80 sets of coal spectral data are enhanced by data augmentation, outlier detection, and dimensional transformation to improve the number and quality of samples. Then, an analytical model combining dilated convolution, multi-level residual connection, and a two-hidden-layer extreme learning machine (TELM), named DR_TELM, was proposed. The model extracted effective features from coal spectral data by a convolutional neural network (CNN) and utilized TELM as a regressor to achieve feature identification and content prediction. The experimental results showed that DR_TELM achieved coefficients of determination (R²) of 0.981, 0.989, 0.990, 0.985, 0.989 and root mean square errors (RMSE) of 0.533, 1.833, 1.111, 1.808, 0.723 for the content prediction of moisture, ash, volatile matter, fixed carbon and higher heating value (HHV), respectively. And while ensuring high accuracy, the test time is only 0.034 s. It is fully demonstrated that DR_TELM can rapidly and accurately analyze coal.

A Deep-Learning-Based Artificial Intelligence System for the Pathology Diagnosis of Uterine Smooth Muscle Tumor

We aimed to develop an artificial intelligence (AI) diagnosis system for uterine smooth muscle tumors (UMTs) by using deep learning. We analyzed the morphological features of UMTs on whole-slide images (233, 108, and 30 digital slides of leiomyosarcomas, leiomyomas, and smooth muscle tumors of uncertain malignant potential stained with hematoxylin and eosin, respectively). Aperio ImageScope software randomly selected ≥10 areas of the total field of view. Pathologists randomly selected a marked region in each section that was no smaller than the total area of 10 high-power fields in which necrotic, vascular, collagenous, and mitotic areas were labeled. We constructed an automatic identification algorithm for cytological atypia and necrosis by using ResNet and constructed an automatic detection algorithm for mitosis by using YOLOv5. A logical evaluation algorithm was then designed to obtain an automatic UMT diagnostic aid that can "study and synthesize" a pathologist's experience. The precision, recall, and F1 index reached more than 0.920. The detection network could accurately detect the mitoses (0.913 precision, 0.893 recall). For the prediction ability, the AI system had a precision of 0.90. An AI-assisted system for diagnosing UMTs in routine practice scenarios is feasible and can improve the accuracy and efficiency of diagnosis.

Machine Learning of Raman Spectroscopy Data for Classifying Cancers: A Review of the Recent Literature

Raman Spectroscopy has long been anticipated to augment clinical decision making, such as classifying oncological samples. Unfortunately, the complexity of Raman data has thus far inhibited their routine use in clinical settings. Traditional machine learning models have been used to help exploit this information, but recent advances in deep learning have the potential to improve the field. However, there are a number of potential pitfalls with both traditional and deep learning models. We conduct a literature review to ascertain the recent machine learning methods used to classify cancers using Raman spectral data. We find that while deep learning models are popular, and ostensibly outperform traditional learning models, there are many methodological considerations which may be leading to an over-estimation of performance; primarily, small sample sizes which compound sub-optimal choices regarding sampling and validation strategies. Amongst several recommendations is a call to collate large benchmark Raman datasets, similar to those that have helped transform digital pathology, which researchers can use to develop and refine deep learning models.

High-Precision Intelligent Cancer Diagnosis Method: 2D Raman Figures Combined with Deep Learning

Raman spectroscopy, as a label-free detection technology, has been widely used in tumor diagnosis. However, most tumor diagnosis procedures utilize multivariate statistical analysis methods for classification, which poses a major bottleneck toward achieving high accuracy. Here, we propose a concept called the two-dimensional (2D) Raman figure combined with convolutional neural network (CNN) to improve the accuracy. Two-dimensional Raman figures can be obtained from four transformation methods: spectral recurrence plot (SRP), spectral Gramian angular field (SGAF), spectral short-time Fourier transform (SSTFT), and spectral Markov transition field (SMTF). Two-dimensional CNN models all yield more than 95% accuracy, which is higher than the PCA-LDA method and the Raman-spectrum-CNN method, indicating that 2D Raman figure inputs combined with CNN may be one reason for gaining excellent performances. Among 2D-CNN models, the main difference is the conversion, where SRP is based on the structure of wavenumber series with the best performances (98.9% accuracy, 99.5% sensitivity, 98.3% specificity), followed by SGAF on the wavenumber series, SSTFT on wavenumber and intensity information, and SMTF on wavenumber position information. The inclusion of external information in the conversion may be another reason for improvement in the accuracy. The excellent capability shows huge potential for tumor diagnosis via 2D Raman figures and may be applied in other spectroscopy analytical fields.