Page 20 - Kaleidoscope Academic Conference Proceedings 2022
P. 20

MINOHEALTH.AI: A CLINICAL EVALUATION OF DEEP LEARNING
                      SYSTEMS FOR THE DIAGNOSIS OF PLEURAL EFFUSION AND
                                CARDIOMEGALY IN GHANA, VIETNAM AND
                                      THE UNITED STATES OF AMERICA



                                                     Darlington Akogo


                                minoHealth AI Labs, karaAgro AI, Runmila AI Institute, Ghana


            A rapid and accurate diagnosis of medical conditions like cardiomegaly and pleural effusion is of the
            utmost  importance  to  reduce  mortality  and  medical  costs,  and  artificial  intelligence  has  shown
            promise  in  diagnosing  medical  conditions.  We  evaluated  how  well  Artificial  Intelligence  (AI)
            systems,  developed  by  minoHealth  AI  Labs,  perform  at  diagnosing  cardiomegaly  and  pleural
            effusion, using chest x-rays from Ghana, Vietnam and the USA, and how well AI systems perform
            when  compared  with  radiologists  working  in  Ghana.  The  evaluation  dataset  used  in  this  study
            contained 100 images randomly selected from three datasets. The deep learning models were further
            tested on a larger Ghanaian dataset containing 561 samples. Two AI systems were then evaluated on
            the evaluation dataset, whilst we also gave the same chest x-ray images within the evaluation dataset
            to four radiologists, with 5 - 20 years’ experience, to give their independent diagnoses.


            For  cardiomegaly,  minoHealth.ai  systems  scored  an  Area  Under  the  Receiver  Operating
            Characteristic Curve (AUC-ROC) of 0.9 and 0.97 while the AUC-ROC of individual radiologists
            ranged from 0.77 to 0.87. For pleural effusion, the minoHealth.ai systems  scored 0.97 and 0.91,
            whereas  individual  radiologists  scored  between  0.75  and  0.86.  On  both  conditions,  the  best
            performing AI model outperforms the best performing radiologist by about 10%. These models will
            be of great use in regions, such as sub-Saharan Africa, where there are few radiologists. They can
            potentially be used to augment the effort of radiologists to improve the diagnosis and treatment of
            chest conditions.









































                                                          – xvi –
   15   16   17   18   19   20   21   22   23   24   25