AIMO Conference

Retinopathy and glaucoma: artificial intelligence is being integrated into ophthalmology clinical pathways

From screening to the monitoring of major conditions, the integration of new technologies and algorithms supports local healthcare provision

by Alberto Piatti*

9 October 2026

Aggiungi Il Sole 24 Ore
ai preferiti su Google

3' min read

Translated by AI

Versione italiana

Key points

Early diagnosis of diabetes complications
The benefits and main challenges
The Aimo Congress

3' min read

Translated by AI

Versione italiana

The impact of artificial intelligence (AI) on clinical practice is now one of the key topics within the international scientific community. Interest in this area within the scientific literature is at an all-time high: suffice it to say that, over the past year, more than 3 million research papers on the use of AI in healthcare have been published in peer-reviewed scientific journals, with as many as 21,200 of these concerning its use in the field of ophthalmology (data from Google Scholar).

To mark World Sight Day, the Italian Association of Ophthalmologists (AIMO) devoted a substantial section of its national congress to analysing this technological revolution, with the aim of establishing practical guidelines for the integration of automated systems into diagnostic, therapeutic and care pathways (PDTA).

Early diagnosis of diabetes complications

One of the most well-established areas of application for AI is the early diagnosis of diabetic eye complications. Screening is carried out on diabetic patients to identify the two most feared complications: retinopathy and diabetic macular oedema. Numerous algorithms have been validated by the EC to recognise the early signs of diabetic retinopathy.

The algorithms process images of the fundus captured using a digital camera, known as a retinograph.

In our clinical practice, within the local Diabetes and Ophthalmology departments of the ASL To5, we have had the opportunity to test and use two different AI programmes which have demonstrated high sensitivity (over 98 per cent) and excellent specificity (91 per cent for referable cases, i.e. those requiring an ophthalmological consultation).

The benefits and main challenges

The systematic adoption of AI in screening programmes offers clear operational advantages: diagnostic accuracy and standardisation, process efficiency, the possibility of using portable devices in decentralised settings, thereby reaching population groups that would otherwise be excluded, the streamlining of waiting lists, and the timely referral of only high-risk cases to a specialist.

Despite the many advantages, the use of AI algorithms also presents a number of challenges. Here are the main ones:

- Image quality: the performance of the algorithms depends on the quality of the images acquired; blurred images or those containing artefacts may compromise diagnostic accuracy.

- Bias in training datasets: if the data used for training is not representative of the target population, the algorithm may not generalise well.

- Medical-legal implications: reliance on automated systems raises legal issues in the event of diagnostic errors.

- Acceptance by healthcare staff: the integration of AI into clinical practice requires a cultural shift and appropriate training.

- Privacy and data security: the handling of patient images and data must comply with privacy regulations and ensure the security of information.

The Aimo Congress

The AIMO National Congress has once again highlighted the use of artificial intelligence to support clinicians. The central role always rests with the ophthalmologist and their relationship with the patient, which is of paramount importance in chronic conditions. At present, we cannot view favourably that aspect of ‘generative’ AI which seeks to propose diagnoses and treatments, thereby replacing the human professional. Screening procedures (such as for diabetic retinopathy) might warrant separate consideration, where an initial assessment by artificial intelligence could speed up processes and enable more widespread screening campaigns. In fact, in Italia – as well as in many other European countries – the uptake of screening for retinal complications in diabetic patients remains low (around 30 per cent of the diabetic population – data from the Annali di AMD).

*ASL TO5 ophthalmologist and AIMO councillor

Brand connect

I prossimi eventi

Tutti gli eventi

Notizie e approfondimenti sugli avvenimenti politici, economici e finanziari.