Multi-Sensor Fusion Approach to Drinking Activity Identification for Improving Fluid Intake Monitoring

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript proposes an improved method combining wrist-worn motion signals and ear microphone acoustic signals to identify drinking activities, showing better results than single-modal approaches. I have several questions regarding the manuscript's methodology.

1. The study's focus on sitting and standing conditions is good, but it's worth considering if walking conditions were studied or could be in the future for a more complete understanding.

2 While the work involves participants of similar ages in sitting and standing positions, it's important to investigate if age differences affect signal.

3. The impact of sitting and standing postures on modeling accuracy needs to be further discussed.

4. Considering the position of sensors on both wrists and its effect on signal acquisition would add depth to the study and improve its applicability.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

The authors propose a system to evaluate the drinking activities which can help the people to improve their fluids intake habits. The system is interesting but some sophisticated and maybe it is difficult to implements in a commun environment. However, if a special study about this topic is required, the propose methodology is adequate, the acquisition and processing data is proposed, considering different scenarios.

On the other side, although the study, the methodology and the results are interesting, is it not clear the novelty. Please could you comment about.

 

The text on the figure 1, the legends on figures 7 to 13 and the tags of the figure 3 are too small, please can you improve them.

 

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

The authors developed a multi-sensor approach incorporating machine learning algorithms to monitor fluid intake. The concept and results are interesting. This reviewer recommends its publication in Applied Sciences after a minor revision to address the minor comments below:

1. Please add the scale bars and the units in Figure 4.

2. Please add quantified results (e.g. recognition accuracy) in the abstract and the conclusion to highlight the work.

3. Can the method identify the size of fluid intake (large and small swallows)? What is the minimum fluid intake that this method can successfully identify? 

4. By using this method, will different fluids lead to different recognition accuracy (e.g. viscosity, sparking, etc.)? 

5. In the discussion section, please talk more specifically about future directions and potential applications of this work.

Author Response

Please see the attachment

Author Response File: Author Response.docx