Diagnostics
Early, precise detection makes it possible to fight cancer and provide tailored therapies. A challenge in modern diagnostics is to provide clinicians with the most accurate information through the least invasive procedures. Accurate diagnostics support personalised treatments that improve the chances of complete remission. The following datasets cover technologies used to extract and analyse tumour samples, and apparatus and agents that support imaging of healthy and tumoural tissues using non-invasive procedures.
Biopsies
Tissue biopsies are still indispensable for the diagnosis of solid tumours. This invasive procedure must be performed in conditions that provide a usable sample to the clinician and reduce post-operative complications. It is essential to prevent any cancer cells spreading during the sample extraction.
Safer sample extraction
The following datasets relate to risk-mitigating technologies to be combined with extraction techniques, e.g. sealing the track after removing the biopsy instruments. Different technologies reduce risks to patients:
Detection of cancer cells in a sample
The following datasets identify patent documents relating to assays and immunoassays performed on biopsies to identify and characterise cancer cells.
Detection of cancer cells by immunoassay
Identification of surface markers of cancer cells by immunoassay
Detection of tumour-infiltrating cells by (immuno)assays
Imaging
Tumour imaging is a non-invasive procedure based on X-ray, nuclear or ultrasonic imaging, often in combination with a labelled or contrast agent that gives precise information on the location and shape of the tumour. Tumour imaging reduces the risks to the patient and facilitates surgery or radiotherapy.
Non-invasive interventions
The same instruments used in early diagnosis can also be used in cancer diagnosis. This dataset concerns cancer diagnosis using ultrasonic, sonic or infrasonic waves.
Artificial intelligence in cancer imaging
The following datasets concern the use of artificial intelligence (AI) to analyse images from a variety of imaging technologies to improve their accuracy and sensitivity in detecting neoplasia or early cancer. They relate to computational methods based on machine learning using the following technologies:
Support vector machines (SVMs)
Convolutional neural networks (CNNs)
General adversarial networks (GANs)
The following dataset relates to the general application of computational methods from the fields of bioinformatics and healthcare informatics, including image-analysis-based methods.
Bioinformatics and healthcare informatics
Apparatus
To accurately screen, stage and provide a 3D image of a tumour, it is essential to have apparatus that has an excellent tumour-tissue-to-background ratio. Clinicians have access to multiple technologies that can support their diagnosis and guide them to the appropriate intervention. The following datasets relate to such apparatus.
Imaging agents
The sensitivity of the detection and the tumour-to-background ratio provided by the apparatus referred to in the previous section can be further improved by use of imaging agents comprising radiolabelled or contrast agents, including biomolecules such as antibodies.