Diabetic foot ulcer detection using deep learning approaches Puneeth N. Thotad a, b , Geeta R. Bharamagoudar c , Basavaraj S. Anami d, * a Department of Master of Computer Applications, KLE Institute of Technology Hubballi, 580027, India b Visvesvaraya Technological University, Jnana Sangama, Belagavi, 590018, India c Department of Computer Science & Engineering, KLE Institute of Technology Hubballi, 580027, India d School of Computer Science and Engineering, KLE Technological University, Hubballi, 580031, India ARTICLE INFO Keywords: Deep learning Convolutional neural network Diabetic foot ulcer Digital healthcare Diabetics ABSTRACT The most recurrent side effect of diabetes is diabetic foot ulcers and if unattended cause imputations. Diabetic feet affect 15% to 25% of diabetic people globally. Diabetes complications are due to less or no awareness of the consequences of diabetes among diabetic patients. Technology leveraging is an attempt to create distinct, affordable, and simple diabetic foot diagnostic strategies for patients and doctors. This work proposes early detection and prognosis of diabetic foot ulcers using the EfficientNet, a deep neural network model. EfficientNet is applied to an image set of 844-foot images, composed of healthy and diabetic ulcer feet. Better performance is obtained compared to earlier models using EfficientNet by carefully balancing network width, depth, and image resolution. The EfficientNet performed better compared to popular models like AlexNet, GoogleNet, VGG16, and VGG19. It gave maximum accuracy, f1-score, recall, and precision of 98.97%, 98%, 98%, and 99%, respectively. 1. Introduction Diabetic foot ulcers (DFUs) are foot injuries and serious cases of diabetes. Reports indicate that there were only 151 million diabetic in- dividuals worldwide in the year 2000, this number increased to over 422 million in 2014 and has been raised to approximately 537 million in 2021. The prevalence of diabetes disease attained an increase of 10.5% among adults over 18 years of age between 2000 and 2021 years. By the end of 2035, the number of diabetic persons is expected to rise to 630 million, as given in Table 1. In addition, 80% of these patients live in developing countries, which lack healthcare facilities and are less aware of patient health conditions [1]. Diabetes foot affects 15% to 25% of these diabetic patients and may face a final stage of foot ulcers which will cause their lower limbs to be amputated, hospitalization of the patient, and finally the death of the diabetic patient when there is no correct treatment [2,3]. Amputation of the foot or limb may occur by the infection of DFUs [4]. The rate of survival is less significant for patients with amputated limbs. It impairs the quality of life, and livelihood, and affects even social participation [5]. Gangrene will be the result of such causes and tissue death due to disease. The burden of diabetes (DFU) seems to increase in the future [6]. Because of the lack of resources and the scarcity of specialists in the treatment of diabetic foot ulcers, more than a million diabetic patients who are at elevated risk of diabetes will lose part of their foot every year. It is observed that for every 20 sec one diabetic foot is operated on. Fig. 1. (a)-(d) presents the healthy & normal foot and Fig. 1. (e)–(h) shows the ulcers on the foot of diabetic patients. A comprehensive analysis of medical data is necessary for pro- fessionals to establish an accurate diagnosis. Traditional diagnostic methods are labor-demanding and prone to human errors. The use of computer-assisted diagnostic procedures lowers costs while enhancing performance. Recent developments in mobile and wearable health de- vices help control diabetes and its consequences by extending remission and improving the quality of life for patients by sensing and controlling harmful foot pressure and inflammation [7]. Sensors are tools that identify physical, chemical, and biological signals and offer a mechanism to quantify and record such signals. Numerous industrial sensor tech- nologies have medical uses. When novel sensors and sensor-dependent * Corresponding author. E-mail address: anami_basu@hotmail.com (B.S. Anami). Production and hosting by Elsevier Contents lists available at ScienceDirect Sensors International journal homepage: www.keaipublishing.com/en/journals/sensors-international https://doi.org/10.1016/j.sintl.2022.100210 Received 9 October 2022; Received in revised form 23 November 2022; Accepted 25 November 2022 Available online 5 December 2022 2666-3511/© 2022 The Authors. Publishing services by Elsevier B.V. on behalf of KeAi Communications Co. Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Sensors International 4 (2023) 100210