Valuation of automatized ASPECTS as an artificial intelligence tool in daily clinical
*Correspondencia: Dra. Selma del Campo Estepar. Departamento de Radiología. Hospital Ribera Povisa. C/ Salamanca, 5. E- 36211 Vigo, Pontevedra.
E-mail: sdelcampo@povisa.es
Aims: To evaluate an automated ASPECTS (ASPECTS-a) software against two radiologists' reading of CT scans requested from the Emergency Department. Describe the most frequent failures of the ASPECTS-a.
Material and methods: All the cranial CT Scans requested by the Emergency Department in one month were collected. The following data were recorded: age, sex, the reason for requesting the study, and imaging findings. A program was used that provides an ASPECTS score automatically. Subsequently, 2 radiologists independently reviewed all of the studies and provided the visual ASPECTS (ASPECTS-v). In case of discrepancy, a new reading was made by consensus.
Results: A total of 295 brain CT scans (45.1% male) with a mean age of 65 ± 20.0 years were included. 91.8% were interpreted as ASPECTS-v 10 in both cerebral hemispheres by both readers. ASPECTS-a scored 45% with ASPECTS 10 in both cerebral hemispheres. In 152 (51.5%) the ASPECTS-a and the ASPECTS-v did not coincide. The causes of the discrepancy were mainly due to segmentation errors (usually due to asymmetric atrophies). Most of the segmentation errors were located in the head of the caudate nucleus, observed in 60 studies.
Conclusions: ASPECTS-a is a powerful and helpful tool, but human supervision is always necessary, particularly in groups of patients with pre-existing brain changes.
Resultados Se realizaron un total de 295 TC cerebrales urgentes con una edad media de 65 ± 20 años. El 91,8% lo interpretaron los dos lectores como ASPECTS 10 en ambos hemisferios cerebrales. El ASPECTS-a puntuó el 45% con ASPECTS 10 en ambos hemisferios cerebrales. En 152 (51,5%), el ASPECTS-a y el ASPECTS-v no coincidieron. Las causas de la discrepancia fueron fundamentalmente por errores en la segmentación (generalmente por atrofias asimétricas). La mayor parte de los errores en la segmentación se localizaban en la cabeza del núcleo caudado, lo que se observó en 60 estudios.
Conclusiones El ASPECTS-a es una herramienta potente y de gran ayuda, pero siempre es necesaria una supervisión humana, particularmente en grupos de pacientes con cambios cerebrales preexistentes.