Themed collection SPEC 2022: International Conference on Clinical Spectroscopy

Aidan D. Meade, Fiona M. Lyng and Hugh J. Byrne introduce the Analyst themed issue on SPEC 2022: International Conference on Clinical Spectroscopy.

Pancreatic cancer, particularly Pancreatic ductal adenocarcinoma, remains a highly lethal form of cancer with limited early diagnosis and treatment options.

Our study assessed the most important light-based markers for surgical and laparoscopic delineation of colorectal cancer margins. Real-time and accurate tissue identification can improve surgical outcomes and obviate the need for multiple biopsies.

This work describes how pre-analytical variables such as sample handling, storage and processing can influence the biochemical signature of lymphocytes.

We compare dimensionality reduction approaches and their effect on identifying cancer in infrared microscopic images. Neural networks trained on reduced spectra perform surprisingly well, indicating the importance of spatial information.

This study developed gelatin-based phantoms optimized across a broad range of optical properties. These well-characterized, multi-biomarker phantoms have the potential to expedite the development of multimodal optical spectroscopy devices for clinical translation.

Spatially offset Raman spectroscopy revealed a correlation of protein and lipid content with non-melanoma skin cancer. Multiple factor analysis on the same data demonstrated clustering based on pathology rather than interpatient variability.

Raman MicroSpectroscopy (RMS) is a powerful label-free tool to probe the effects of drugs at a cellular/subcellular level.

Spectra taken using a HWN Raman probe from bisected mastectomy specimens are capable of identifying tumour versus normal tissue through the difference in water, protein and lipid content.

Fusion of FTIR with H&E images reveals the contributions of collagen, DNA and cytokeratin in metastatic/lymphoid tissue at high spatial resolution.

Raman optical tweezers identifies the invasive bladder cancer cell line, strengthening the argument for a spectroscopic based urine cytology test.

Raman spectroscopic technology for quick and accurate assessment of tumor resection margins.

High-density, compact self-assembled gold nanostar films were prepared by convective self-assembly. An interplay between the enhanced Raman scattering and emission of fluorophores points toward their SERS/SEF dual detection capability.

Extracellular vesicles (EVs) extracted from human bile could be versatile biomarkers for the identification of gall bladder cancer.

Raman spectroscopy approach based on dried droplets, for mixed UTI bacterial pathogens identification.

A novel weakly supervised anomaly detection machine learning method is used to identify non-normal tissue subgroups, interferents, and other tissue components in IR Micro-spectroscopy liver tissue data when trained on a single control group of data.

Using a combination of vibrational and chiroptical spectroscopy together with principal component analysis and a random forest algorithm, we have successfully distinguished patients with hepatocellular carcinoma from patients with cirrhosis.

In vivo Raman spectroscopy coupled with machine learning methods was used for tissue diagnostics of precancerous and cancerous lesions of the colon and rectum.

Comparison of (a) H&E image of oral squamous cell carcinoma with (b) ratio of absorbance at 1252 cm⁻¹/1285 cm⁻¹ and (c) pseudo-coloured labelled image of seven types of tissue found in oral tissue.

A clinical prototype Raman macroscopic imaging system for machine-learning-based molecular tissue margin characterization over a large field of view.

Ex situ biofilm samples were used to study the interactions between SERS-enhancing Ag and Au particles and biofilm components without the complexity associated with highly structured in situ biofilms.

Intestinal inflammation-driving T cells within the epithelial layer of the small intestine can be identified and further characterized by Raman spectroscopy ex vivo.

Raman spectroscopy (RS) detects the quality of lipoproteins (LPs) extracted from obese patients and healthy subjects. RS simultaneously identified clear differences in the LPs composition and in the oxidation state between the two groups.

A recurrent neural network trained on time domain data can accurately identify brain tumours from serum spectral data.

We combined infrared (IR) microscopy of bladder sections with machine learning (partial least squares-discriminant analysis) to provide an automated methodology for the prediction of the presence of cancer, invasiveness and grade.