Abnormal Gait and Tremor Detection in the Elderly Ambulatory Behavior Using an IoT Smart Cane Device

Round 1

Reviewer 1 Report

Paper deals with important task. The authors proposed a novel approach for abnormal gait and tremor detection using smart walking cane.

Suggestions:

1.       It would be good to add clear point-by-point the main contributions at the end of the Introduction section

2.       The related works section should be extended using disadvantages of the existing methods

3.       The authors should add all optimal parameters for all investigated methods

4.       There are no numerical resuts for isolated forest and SVM

5.       There are no srong Results and Discussion sections. Please add it

6.       The conclusion section should be extended using: 1) numerical results obtained in the paper; 2) limitations of the proposed approach; 3) prospects for future research.

Please add more references.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments to the authors:

Marion O. Adebiyi and colleagues evaluated the use of a smart walking cane for the detection of abnormal walking patterns. They constructed a prototype that captured the movements of the cane using accelerometer/gyroscope, GSM, and GPS signals. The authors simulated different walking patterns and used PCA and LDA to classify normal vs. abnormal walking pattern.

The authors should be congratulated on the novelty of this approach and on performing these complex studies. Although this reviewer is very enthusiastic about this work, the manuscript is difficult to follow and underdeveloped overall.

Major comments:

1.      This reviewer believes that the manuscript in its present form needs some restructuring. A large portion of section “3. Tremor Detection” should go into “Materials and methods”. The technical and statistical descriptions in “5. Anomaly Detection in Gait Ambulation” should also go into “Materials and methods”.

2.      The methods section should contain enough detail (e.g., pertaining to the construction of the cane, the names of the accelerator chip, etc.) to allow future researchers to reproduce the same results.

3.      The methods for simulating different walking patterns by the three individuals need more detail (e.g., how were they simulated, how slow is slow walking, how is vibrating different from shaking arm or rhythmic wrist movements, how much data were captured for each of the six categories, how much data was used for training and testing)

4.      The authors need to further elaborate on the results section of the manuscript. The authors mentioned that the aim of the manuscript was to detect “tremor and gait abnormality”. However, the results only showed discrimination between training/test data vs. “anomaly observation”. What is “anomaly observation”? Does it contain all the six categories/patterns of walk? If so, the number of anomalous data seems too low to capture all the five non-normal walking patterns.

Minor comments:

1.      Line 5, introduction: “2523 billion worldwide population of older people by 2050”. Please double-check the number of people (it should not be 2523 billion).

2.      Second paragraph, introduction: the author mentioned that “Our approach has so many advantages over all other existing methods.” The authors should reference/cite some of those existing methods.

3.      Introduction, section “(B). Elderly Activities Evaluation”: most of the statements in this section need proper references, e.g., “Some studies have it that…”.

4.      Figures (e.g., Figure 3) need proper labeling of x- an y- axes.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors did a lot of works. Paper can be accepted

Reviewer 2 Report

This reviewer believes that the manuscript has been sufficiently improved and thanks the authors for the corrections!