Although artificial intelligence (A.I.) has experienced rapid growth over the past few years, the term A.I. was first coined by John McCarthy (1) in 1955. A.I. is defined as the ability of machines to learn and display intelligence. Different experts (2) considered five levels of automation in medicine.

Let’s talk about each level in sperm analysis field:

  • Human only, no A.I. involved: This is the standard method (and still a lot of laboratories use it) where the technician use a microscope to check manually the different semen parameters. This is the most subjective and with more variation between different technicians.
  • Shadow mode: The technician performs the analysis manually but save all the results (and sometimes images) in a database. This is the first step to have traceability and data to train A.I. systems.
  • A.I. assistance: Here is where appears the Computer-aided semen analysis systems (also known as CASA). The system performs the analysis automatically and the technician can check and if necessary, modify the analysis. Still is necessary a well-trained technician and a lot of manual steps. Microptic developed the Sperm Class Analyzer (SCA), a CASA system with this type of A.I., in 1997 and today is one of the most used around the world.
  • Partial Automation: With partial automation, an A.I. system can come up with its own diagnosis; but if it is not confident enough about it, the A.I. turns to physicians for help. Microptic developed the first fully automatic CASA system for different semen parameters in 2018. The name of this system is “SCA SCOPE” and it uses different types of A.I. like deep learning and machine learning. To use this system is not necessary a well-trained technician, the system performs all steps automatically and is just necessary the final step of validation by a trained technician.
  • Full automation: As the name suggests, full automation processes are performed by an A.I. alone and do not involve human input. As semen is a very heterogeneous fluid and in medicine is always necessary to be 100% sure about the diagnostic, it is nearly impossible that in the future will exist a semen analysis system in this level.

The 5 levels of automation

The 5 levels of automation infographic. (Source: medicalfuturist.com)

A.I. technology has been applied in semen analysis field for many years and today is nearly impossible to think that somebody can perform a sperm analysis without a CASA system.

In my opinion, it is necessary to use automate systems to obtain objective results. The same way that doctors use a thermometer to know exactly if somebody has fever or not, the same way police use speed cameras to know if a car exceed the limit, spermatologists must use an automate system to get repeatable results, does not matter if the analysis is done by an operator or another, in the morning or in the evening, on a quiet day or with many samples. Systems with cutting edge A.I. technology are accurate, easy to use and always possible to validate if results are correct or not.

So, the title of this blog is incomplete. The best title is:
Artificial intelligence will not replace spermatologists.

However, spermatologists who use A.I. will replace those who do not.

Eduard Sànchez Mallol
CEO, Microptic

References

(1) McCarthy, John; Minsky, Marvin; Rochester, Nathan; Shannon, Claude (1955), A Proposal for the Dartmouth Summer Research Project on Artificial Intelligence

(2) Danielle S Bitterman; Hugo J W L Aerts; Raymond H Mak; Lancer digit Health (2020). Approaching autonomy in medical artificial intelligence

Subscribe to Directory
Write an Article

Recent News

Exposure to Heat and Cold During Pregnan...

The research team observed changes in head circumf...

Using mobile RNAs to improve Nitrogen a...

AtCDF3 gene induced greater production of sugars a...

El diagnóstico genético neonatal mejor...

Un estudio con datos de los últimos 35 años, ind...

Highlight

Eosinófilos. ¿Qué significa tener val...

by Labo'Life

​En nuestro post hablamos sobre este interesante tipo de célula del...

A new tool to predict nucleosome positio...

by IRB Barcelona - Instituto de Investigación Biomédica

A team of researchers at IRB Barcelona have developed an innovative co...

Photos Stream