“Biophotonics is a key technology for gastroenterological diagnostics,” said Maximilian Waldner, Professor at the Erlangen University Clinic, at the “PHOTONICS 4.0—Optical Health Technologies” forum in Berlin recently. Experts from the fields of medicine and photonics were invited to the event by the German industry association SPECTARIS to explore the opportunities offered by increasingly connected, optical health technologies.
For decades now, endoscopes have been used in gastroenterology in order to identify inflammations and tumors in the esophagus, stomach and bowel. According to Waldner, however, there are a number of problems. For example, it is difficult to tell pathological tissue changes from those that are harmless during conventional endoscopies. An exact diagnosis is only possible with tissue samples, which take time and often cause complications. In many cases, it takes days for the laboratory findings to come back. This means a distressing wait for patients and lost time in the fight against diseases such as bowel cancer.
Every year, 70,000 people in Germany develop colorectal cancer and 30,000 patients die from it. According to Waldner, deaths could be prevented in many cases if the mutations were detected earlier and removed during endoscopies. However, studies show that tumors in their early stages are often missed. Statistics reveal that one in thirteen colorectal cancers remains undetected during screening endoscopies. Diagnostic systems that provide doctors with more targeted support would therefore be ideal. “For example, red-flag technologies during an endoscopy which clearly show where the doctor needs to look more closely,” said Waldner. However, approaches such as high-definition or chromoendoscopy, where dyes are used to provide clear contrasts on mucous membranes, as well as procedures which rely on optical filters and digital data processing have failed to produce truly impressive results so far.
Doctors are pinning their hopes on technologies that broaden their field of vision and make it easier to navigate the intestines with all their twists and turns. Examples include the Third Eye Retroscope from Avantis Medical Systems, Full Spectrum Endoscopy from the U.S. company Boston Scientific or molecular endoscopy where markers in the form of antibodies or peptides are used. These make mutated tumor tissue visible through fluorescence. However, the quality of the diagnoses depends on the quality of the markers.
The aim of the development should be “in-vivo histology” on the basis of which doctors can decide immediately whether they need to remove tissue. Confocal laser endomicroscopy is one possibility. A confocal miniature microscope integrated into the endoscope enables high-resolution microscopic images to be produced inside the body. Lasers light up the tissue for this purpose. “Lateral resolutions of less than 1 micrometer and a penetration depth of up to 150 micrometers help us to distinguish healthy mucous membrane structures from pathologically mutated ones,” said Waldner. However, the method is complex because dye needs to be applied beforehand. There is also a lack of approved dyes. One alternative is multi-photon microscopy which makes tissue structures visible with the help of non-linear optical effects—without dyes. However, the endoscopes required are not yet available on the market.
Like the gastroenterologist from Erlangen, various doctors used the SPECTARIS forum to demonstrate the current possibilities and limits of biophotonic diagnostics. It is clear that the number of procedures will increase significantly. With optical coherence tomography, fluorescence lifetime imaging microscopy (FLIM), raman spectroscopy, raman imaging and coherent anti-stokes raman spectroscopy (CARS), various new methods are making their way onto the market. Even opto-acoustic procedures for non-invasive diagnostics are becoming feasible.
This variety is both a blessing and a curse. In the future, an entirely new level of prevention could be possible if multimodal approaches can be introduced in practice. Patients could undergo a wide range of optical diagnostic procedures during one examination. If the findings are suspicious, mutated tissue could be removed immediately before metastases form. The image data of healthy patients could be used as reference data for each subsequent diagnosis. If this vision is to become reality, doctors need more help when evaluating images. Automated diagnostics would be possible if learning systems can train using large quantities of real patient data. They could learn to scan tissue systematically for even the smallest of changes and assess these changes by comparing them with the disease progression of earlier patients. However, the forum in Berlin showed that this is barely feasible. After all, the image data from real-life clinical situations is not available and a database of this type cannot be set up for data privacy reasons. Automated photonics 4.0 diagnostics will have to find other ways of training software to distinguish between diseased and healthy tissue.