Quantum Cascade Laser-Based Infrared Imaging as a Label-Free and Automated Approach to Determine Mutations in Lung Adenocarcinoma

Infrared Imaging Combined with Machine Learning for Detection of the (Pre)Invasive Pancreatic Neoplasia

With the challenge of limited early stage detection and a resulting five-year survival rate of only 13%, pancreatic ductal adenocarcinoma (PDAC) remains one of the most lethal cancers. Replacing the high-cost and time-consuming grading of pancreatic samples by pathologists with automated diagnostic approaches can revolutionize PDAC detection and thus accelerate patient admission into the clinical setting for treatment. To address this unmet diagnostic need and facilitate the shift of tissue screening toward automated systems, we combined stain-free histology-specifically, Fourier-transform infrared (FT-IR) imaging-with machine learning. The obtained stain-free model was trained to distinguish between normal, benign, and malignant areas in analyzed specimens using hematoxylin and eosin stained pancreatic tissues isolated from KC (KrasG12D/+; Pdx1-Cre) or KPC mice (KrasG12D/+; Trp53R172H/+; Pdx1-Cre). Due to the pancreas-specific mosaic expression of the mutant Kras and Trp53 genes, changes in pancreatic tissues of this mouse model of PDAC closely mirror the gradual transformation of normal pancreatic epithelia into (pre)malignant structures. Thus, this mouse model provides a reliable representation of human disease progression, which we tracked in our study with a Random Forest classifier to achieve accurate detection at the cellular level. This approach yielded a comprehensive model that distinguishes normal pancreatic tissues from pathological features such as pancreatic intraepithelial neoplasia (PanIN), cancerous regions, hemorrhages, and collagen fibers, as well as a streamlined model designed to rapidly identify normal tissues versus pathologically altered regions, including PanINs. These models offer highly accurate diagnostic tools for the early detection of pancreatic malignancies, thus significantly improving the chance for timely therapeutic intervention against PDAC.

Label-free Aβ plaque detection in Alzheimer's disease brain tissue using infrared microscopy and neural networks

We present a novel method for the label-free detection of amyloid-beta (Aβ) plaques, the key hallmark of Alzheimer's disease, in human brain tissue sections. Conventionally, immunohistochemistry (IHC) is employed for the characterization of Aβ plaques, hindering subsequent analysis. Here, a semi-supervised convolutional neural network (CNN) is trained to detect Aβ plaques in quantum cascade laser infrared (QCL-IR) microscopy images. Laser microdissection (LMD) is then used to precisely extract plaques from snap-frozen, unstained tissue sections. Mass spectrometry-based proteomics reveals a loss of soluble proteins in IHC stained samples. Our method prevents this loss and provides a novel tool that expands the scope of molecular analysis methods to chemically native plaques. Insight into soluble plaque components will complement our understanding of plaques and their role in Alzheimer's disease.

Dimensionality reduction for deep learning in infrared microscopy: a comparative computational survey

While infrared microscopy provides molecular information at spatial resolution in a label-free manner, exploiting both spatial and molecular information for classifying the disease status of tissue samples constitutes a major challenge. One strategy to mitigate this problem is to embed high-dimensional pixel spectra in lower dimensions, aiming to preserve molecular information in a more compact manner, which reduces the amount of data and promises to make subsequent disease classification more accessible for machine learning procedures. In this study, we compare several dimensionality reduction approaches and their effect on identifying cancer in the context of a colon carcinoma study. We observe surprisingly small differences between convolutional neural networks trained on dimensionality reduced spectra compared to utilizing full spectra, indicating a clear tendency of the convolutional networks to focus on spatial rather than spectral information for classifying disease status.

Digital Histopathology by Infrared Spectroscopic Imaging

Infrared (IR) spectroscopic imaging records spatially resolved molecular vibrational spectra, enabling a comprehensive measurement of the chemical makeup and heterogeneity of biological tissues. Combining this novel contrast mechanism in microscopy with the use of artificial intelligence can transform the practice of histopathology, which currently relies largely on human examination of morphologic patterns within stained tissue. First, this review summarizes IR imaging instrumentation especially suited to histopathology, analyses of its performance, and major trends. Second, an overview of data processing methods and application of machine learning is given, with an emphasis on the emerging use of deep learning. Third, a discussion on workflows in pathology is provided, with four categories proposed based on the complexity of methods and the analytical performance needed. Last, a set of guidelines, termed experimental and analytical specifications for spectroscopic imaging in histopathology, are proposed to help standardize the diversity of approaches in this emerging area.

Fast and label-free automated detection of microsatellite status in early colon cancer using artificial intelligence integrated infrared imaging

PURPOSE

Microsatellite instability (MSI) due to mismatch repair (MMR) defects accounts for 15-20% of colon cancers (CC). MSI testing is currently standard of care in CC with immunohistochemistry of the four MMR proteins representing the gold standard. Instead, label-free quantum cascade laser (QCL) based infrared (IR) imaging combined with artificial intelligence (AI) may classify MSI/microsatellite stability (MSS) in unstained tissue sections user-independently and tissue preserving.

METHODS

Paraffin-embedded unstained tissue sections of early CC from patients participating in the multicentre AIO ColoPredict Plus (CPP) 2.0 registry were analysed after dividing into three groups (training, test, and validation). IR images of tissue sections using QCL-IR microscopes were classified by AI (convolutional neural networks [CNN]) using a two-step approach. The first CNN (modified U-Net) detected areas of cancer while the second CNN (VGG-Net) classified MSI/MSS. End-points were area under receiver operating characteristic (AUROC) and area under precision recall curve (AUPRC).

RESULTS

The cancer detection in the first step was based on 629 patients (train n = 273, test n = 138, and validation n = 218). Resulting classification AUROC was 1.0 for the validation dataset. The second step classifying MSI/MSS was performed on 547 patients (train n = 331, test n = 69, and validation n = 147) reaching AUROC and AUPRC of 0.9 and 0.74, respectively, for the validation cohort.

CONCLUSION

Our novel label-free digital pathology approach accurately and rapidly classifies MSI vs. MSS. The tissue sections analysed were not processed leaving the sample unmodified for subsequent analyses. Our approach demonstrates an AI-based decision support tool potentially driving improved patient stratification and precision oncology in the future.

Shifting Gears in Precision Oncology-Challenges and Opportunities of Integrative Data Analysis

For decades, research relating to modification of host immunity towards antitumor response activation has been ongoing, with the breakthrough discovery of immune-checkpoint blockers. Several biomarkers with potential predictive value have been reported in recent studies for these novel therapies. However, with the plethora of therapeutic options existing for a given cancer entity, modern oncology is now being confronted with multifactorial interpretation to devise “the best therapy” for the individual patient. Into the bargain come the multiverse guidelines for established and emerging diagnostic biomarkers, as well as the complex interplay between cancer cells and tumor microenvironment, provoking immense challenges in the therapy decision-making process. Through this review, we present various molecular diagnostic modalities and techniques, such as genomics, immunohistochemistry and quantitative image analysis, which have the potential of becoming powerful tools in the development of an optimal treatment regime when analogized with patient characteristics. We will summarize the underlying complexities of these methods and shed light upon the necessary considerations and requirements for data integration. It is our hope to provide compelling evidence to emphasize on the need for inclusion of integrative data analysis in modern cancer therapy, and thereupon paving a path towards precision medicine and better patient outcomes.