Introduction/Objective: Machine learning is playing an increasingly important role in precision agriculture. We conducted a study using a collection of 3-channel images with 500 x 500 (910 total images) to predict the most possible classification of carbon levels in soil samples from its cellphone images. Methods: Building upon previous experiments using regression and classification models, we conducted new experiments with deep learning CNN models (AlexNet, VGG, and Resnet). Results: There was improvement of accuracy from 50% to 75% in more recent models. Conclusions: There is currently inconclusive evidence of mapping soil properties to images, provided that the images were very similar in textures / colors. Finally, we confirmed the possibility of enhancing predictions using ML. In addition, we present a path of a predictive ML model in future research with the use of semi supervised learning model such as meta pseudo labels.

It all begins with an idea.

Failure to adopt an interoperable eHealth system limits the accurate communication exchange of pertinent health-care-related data for diagnosis and treatment. Patient data are located in disparate health information systems, and the adoption of an interoperable eHealth system is complex and requires strategic planning by senior health care IT leaders. Grounded in DeLone and McLean’s information system success model, the purpose of this qualitative case study was to explore strategies used by some senior information technology (IT) health care leaders in the successful adoption of an eHealth system. The participants were 8 senior health care IT leaders in the eastern United States who successfully adopted an interoperable eHealth system. Data were collected using semistructured interviews following Kallio’s five phase interview guide and analyzed using thematic analysis. Six themes emerged: eHealth ecosystem, implementation approach, quality, strategy, use/intent to use, and user satisfaction. A key recommendation from results indicates that further identification and development of strategies based on the DeLone and McLean IS success model might benefit successful eHealth adoption and implementation. Positive social change implications include the potential for senior health care IT leaders to identify a framework to enhance accuracy among eHealth systems to reduce medical errors and improve patient care.