| by admin | No comments

Machines can see things we cannot see

From the beginning of time, medicine was based on evidence. Even if societies have changed their approach towards illness and disease from ancient times until the present, one thing remains valid: the ability to see the the pathological process that underlies the condition is an asset for treating it. 

At first, doctors could only use the five basic human senses: touch, sight, hearing, smell and taste; and for a long time, these methods gave good results. In the meantime, medicine has changed in practice, as we discovered that diseases have become more complex. We understood that certain symptoms appear due to changes inside the body that we fail to see with the naked eye. Something had to be done. And it happened. 

First look inside

The X-ray was discovered in 1895, by German physics professor Wilhelm Conrad Röntgen, who found that this type of radiation could penetrate through human tissue but not through bone or metal. The radiation was referred as “X” by Röntgen to denote that it was an unknown sort of energy. For his discovery, the professor was awarded the first Nobel Prize in Physics.  

Although image capture and visualizations have improved over the years, the principles of the technique are the same. Today, like in 1895, conventional radiographic images (usually shortened to x-rays) are produced by a combination of ionizing radiation and light striking a photosensitive surface, which, in turn, produces an image that is subsequently processed. At first, the processing of the film was carried out in a darkroom containing trays with various chemicals and the obtained results were hung up to dry.  

Digital radiography emerged as the replacement of photographic film with a sensitive cassette or plate that is processed by an electronic reader, resulting an digital image. This electronic processing eliminated the need for a darkroom to develop the film or a big storage space to save it. Countless images could be stored on one hard disk or a computer server. More importantly, the photos could be accessed by anybody, at anytime, anywhere in the world. 

By passing x-rays through the human body, traditional radiography can show only five basic densities, which are listed below in order of least to most dense: 

  • Air – which appears blackest on a radiograph.  
  • Fat – which appears in a lighter shade of gray than air. 
  • Soft tissue or fluid – because both soft tissue and fluid appear the same on conventional radiographs, it’s impossible to distinguish the heart muscle from the blood inside the heart.
  • Calcium – usually found within bones. 
  • Metal – which appears whitest on a radiograph. 

News of this discovery spread worldwide and, within a year, doctors from Europe and the United States were using X-rays to locate gun shots, bone fractures, kidney stones and swallowed objects.  Over time, the range of pathologies has widened, including lung and heart illnesses like pneumonia, lung cancer, heart hypertrophy, aneurysms, congenital heart diseases. 

The newer brother of radiology

After a while, CT (or “CAT”) scanners, innovated the medical scene. First introduced in the 1970s, it brought a quantum leap to medical imaging. Using a gantry with a rotating x-ray beam and multiple detectors, a large number of 2D image slices (each of which is millimeters in size) can be formatted in multiple planes.  

A CT scanner, like X-rays, is connected to a computer that processes the data using various algorithms to provide diagnostic-quality images. A CT image is made up of thousands of tiny squares, known as pixels, each of which is assigned a number ranging from -1000 to +1000 in Hounsfield units (HUs) by the computer. 

This sensitivity of capturing an image allowed doctors to easily differentiate one biological structure from another. Doctors were able to see exactly what are the limits of an organ and what happens inside a denser one, which was invisible on radiography until then. Once again, technology has broadened our diagnostic horizons, now being able to detect diseases such as benign and malignant tumors, blood clots, brain and spinal cord injuries, intestinal disorders such as Chron’s disease and internal bleeding. 

A new addition

In 1977 was performed the first magnetic resonance imaging (MRI) exam on a live human patient. The life-saving medical technique had its foundations in the work of physicist Isidor Isaac Rabi, who developed a method of measuring magnetic properties of atomic nuclei, during the 1930s. 

MRI utilizes the potential energy stored in the body’s hydrogen atoms. The atoms are manipulated by strong magnetic fields and radiofrequency pulses. This produces enough tissue-specific energy that allows highly sophisticated computer programs to generate 3D images. 

MRI scanners are not as common as CT. They’re expensive to buy and set up, and they require meticulous site preparation to work effectively. They also have a rather high continuous running cost in general. The machine uses ionizing radiation and can provide far more contrast between different types of soft tissues than CT. MRI is commonly used in neurologic imaging, and it is very sensitive for the detection of soft tissues like muscles, tendons, and ligaments. 

This method of medical imaging can be used to detect brain tumors, traumatic brain injury, congenital anomalies, multiple sclerosis, stroke, dementia, inflammation, and the origins of migraines. It can detect even changes in cartilage and bone structure caused by aging as it is used to diagnose herniated discs, pinched nerves, spinal tumors, spinal cord compression, and fractures. 

The great beyond

XVision develops a software platform that is using Artificial Intelligence to help radiologists analyze medical images faster and more efficiently. Our AI-based algorithms are seamlessly integrated into the doctor’s workflow, helping them analyze chest x-rays and lung CTs up to 25% more efficiently. 

Our team is dedicated to creating the best digital radiology system for medical imaging interpretation. XVision offering includes pathologies detection, localization, and automatic measurements, all of them sent directly to the hospital PACS system.  

XVision In-house created algorithms cover all of a clinician’s most common requirements when it comes to chest image analysis and pulmonary nodules evaluation. Our solution is built around the radiologist’s everyday workflow to increase productivity and efficiency. The Machine Learning operations can enable doctors to identify critical work and focus on the most vital activities by allowing them to easily monitor workloads and processes. 

We make medical imaging technologies better, by providing disruptive solutions like bone suppression and subtraction, automatic calculation of the cardio-thorax index, COVID-19 lesion quantification on lung CTs, pathologies localization and detection, triage and prioritization, pulmonary nodules identification and measurements. All of them powered by AI.  

We are XVision

You can also read about how XVision was used in a national screening program for tuberculosis by clicking here.

Leave a Reply