Computer Vision Techniques for Growth Prediction: A Prisma-Based Systematic Literature Review
Round 1
Reviewer 1 Report
This paper presents a literature review for biological growth prediction using ML and non-ML algorithms. Overall, this paper is structurally sound and fairly easy to follow. But there are several issues this work failed to address properly.
First, in the related work section, several existing literature reviews have been mentioned and summarized, but there are no discussion on how the proposed work differs from those existing work. Additionally, what are the major contributions of this work compared to other existing work. Without those discussion, it is difficult to know if the contribution is justified.
Second, the motivation of this review is weak. In the introduction section ,there is brief mention of the difficulty in acquiring images of living animals consistently and no existing study on machine learning to predict the growth of antlers of deer and cattle and more. But there are no reasoning of how such study would be important and what would be the benefits of having such study. Without proper motivation, it diminish the contribution of this literature review.
Third, the comparison and discussion on current state of research as well as future trend are lacking. There are short discussion in the Conclusion and Future Directions section, but that's overly simplify and almost provide no insights into current state of the research and the possible future research trend or direction. As a literature review, it is important to include those discussion and comparison to provide better insights for other researchers in the field.
Author Response
Thank you for providing us with your valuable feedback and giving us the opportunity to make revisions. We have addressed your comments as follows:
- First, in the related work section, several existing literature reviews have been mentioned and summarized, but there are no discussion on how the proposed work differs from those existing work. Additionally, what are the major contributions of this work compared to other existing work. Without those discussion, it is difficult to know if the contribution is justified.
Answer: We added a new section titled "4. Discussion 4.1. Comparison with Previous Studies and Limitations" to compare our study with the four review studies on growth prediction for various organisms that we referenced. (Reference: page 16 Line 272 - page 17 Line 300)
- Second, the motivation of this review is weak. In the introduction section ,there is brief mention of the difficulty in acquiring images of living animals consistently and no existing study on machine learning to predict the growth of antlers of deer and cattle and more. But there are no reasoning of how such study would be important and what would be the benefits of having such study. Without proper motivation, it diminish the contribution of this literature review.
Answer: As you pointed out, the motivation behind our study was not clear in the previous version. Therefore, we added a new section titled "2. Materials and Methods 2.1. Objective" to explicitly describe the motivation behind our study. (Reference: page 2 Line 52 - page 2 Line 71)
- Third, the comparison and discussion on current state of research as well as future trend are lacking. There are short discussion in the Conclusion and Future Directions section, but that's overly simplify and almost provide no insights into current state of the research and the possible future research trend or direction. As a literature review, it is important to include those discussion and comparison to provide better insights for other researchers in the field.
Answer: We added a new section titled "4. Discussion 4.3. Implications for Future Research" to provide research prospects based on the algorithm in the reviewed papers and potential applications to animal individual identification. (Reference: page 18 Line 336 - page 18 Line 372)
Thank you again for your helpful feedback.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The author has done a good review of computer vision techniques.
However, if the following suggestions are included, the article would be greatly improved.
1. A separate objective section ought to be included in any research paper so that readers can
easily see what the article's goal is. Authors ought to create a separate section for it (though it is a survey).
2. Organization of the paper should be added. (at the end of the Introduction)
3. Since it's a survey article, the readers expect an in-depth technique comparative analysis which is missing. To be more precise, the performance of these techniques should be compared. The authors have mentioned accuracy in table 5 but there are more metrics apart from accuracy for consideration.
4. Authors are considering LSTM, GAN kinds of ML techniques. Hence I recommend few latest articles based on these algorithms be cited.
10.1109/ICTACS56270.2022.9988505
https://doi.org/10.1155/2022/5906877
https://doi.org/10.1016/j.knosys.2021.107994
5. Authors have claimed that they have stuck to PRIMSA 2020 guidelines but still emphasise study selection, more heterogeneity study and limitations in the discussion all are missing.
6. Diagrammatic / Graphical analysis is the most attractive factor for any review work which interests the reader more. Authors should consider including those as a separate analysis section.
7. Authors are advised to follow the IMRAD format for the entire paper.
Author Response
Thank you very much for your thorough review and providing us with the opportunity to make revisions. Please find below our response to your comments, which includes a description of the revised sections and their corresponding references:
- A separate objective section ought to be included in any research paper so that readers can easily see what the article's goal is. Authors ought to create a separate section for it (though it is a survey).
Answer: As you pointed out, the motivation behind our study was not clear in the previous version. Therefore, we added a new section titled "2. Materials and Methods 2.1. Objective" to explicitly describe the motivation behind our study. (Reference: page 2 Line 52 - page 2 Line 71)
- Organization of the paper should be added. (at the end of the Introduction)
Answer: As you suggested, we have added a section to the end of Section 1 to outline the structure of our paper. (Reference: page 2 Line 44 - page 2 Line 51)
- Since it's a survey article, the readers expect an in-depth technique comparative analysis which is missing. To be more precise, the performance of these techniques should be compared. The authors have mentioned accuracy in table 5 but there are more metrics apart from accuracy for consideration.
Answer: Due to the wide variety of research on biological growth covered in these papers, each with its own objectives and data sets, making quantitative comparisons of technical performance is difficult. However, taking into account the lack of deep analysis in the reviewed papers, we added a new section titled "4. Discussion 4.3. Implications for Future Research," in which we discussed future research directions based on the characteristics of the algorithms used. (Reference: page 18 Line 336 - page 18 Line 372)
- Authors are considering LSTM, GAN kinds of ML techniques. Hence I recommend few latest articles based on these algorithms be cited.
10.1109/ICTACS56270.2022.9988505
https://doi.org/10.1155/2022/5906877
https://doi.org/10.1016/j.knosys.2021.107994
Answer: Thank you for suggesting additional references. We have added the reference you provided to section 3.4.1. LSTM (Long Short-Term Memory). (Reference: page 14 Line 227)
- Authors have claimed that they have stuck to PRIMSA 2020 guidelines but still emphasise study selection, more heterogeneity study and limitations in the discussion all are missing.
Answer: To address the limitations of our study, we added a new section titled "4. Discussion 4.1. Comparison with Previous Studies and Limitations," in which we described the limitations of our study by comparing it with existing review papers. (Reference: page 16 Line 272 - page 17 Line 300)
- Diagrammatic / Graphical analysis is the most attractive factor for any review work which interests the reader more. Authors should consider including those as a separate analysis section.
Answer: Thank you for your comments. We have taken your comments into consideration but did not make changes such as moving only the figures of the analysis results to another section, because the research questions and results need to be addressed. However, we created a new section "4. Discussion 4.2. Analysis" to show the graphical results, and added a figure that evaluates the quality of ML papers by field to improve visibility. (Reference: page 17 Line 301 - page 18 Line 335, and page 19)
- Authors are advised to follow the IMRAD format for the entire paper.
Answer: We have revised the structure of our paper to follow the IMRAD format and added a "4. Discussion" section. (Reference: page 16 Line 272 - page 18 Line 372)
Thank you again for your helpful feedback.
Author Response File: Author Response.pdf
Reviewer 3 Report
The authors describe a systematic way in which they have selected the set of papers with the subject "growth prediction". Choosing the set of papers for literature review in this manner is an important step in performing a literature review. However, the literature review itself was not conducted in the paper. It is also not clear which growth type and on which level. The spectrum of growth is very wide starting from the growth of cells (microscopic level) to animals living in the wild. The study is unclear about which type of growth the authors are focusing on. There was no explanation of how technology can contribute to predicting growth and how the different technologies compare in terms of performance.
Author Response
The authors describe a systematic way in which they have selected the set of papers with the subject "growth prediction". Choosing the set of papers for literature review in this manner is an important step in performing a literature review.
Response 1: Thank you very much for time and consideration given to our article. We highly appreciate your feedback on our systematic literature review approach. We agree that this is an important step in conducting a thorough review of the relevant literature. We believe that our approach has allowed us to identify a comprehensive set of papers that are highly relevant to our research question.
However, the literature review itself was not conducted in the paper. It is also not clear which growth type and on which level. The spectrum of growth is very wide starting from the growth of cells (microscopic level) to animals living in the wild. The study is unclear about which type of growth the authors are focusing on. There was no explanation of how technology can contribute to predicting growth and how the different technologies compare in terms of performance.
Response 2: We greatly appreciate your feedback and certainly incorporate your suggestion to provide more clarity on the type of growth that is the important research issue of our article. To address this, we will specify the level of growth on which we are focusing on and explain how various technologies can be leveraged to predict it. To this end, we have included an analytical review of this topic on page 18 line 337 to page 20 line 427. Please refer to the additional note below for further details.
We have understood that growth prediction is feasible at various levels, such as the cellular level (Tumor cellularity), tissue and organismal level (Plant growth), tissue level (Vessel growth), and organismal level (Bone morphology) in Sections 3.1 and 3.2.
To predict changes in deer antler growth, such as the growth and replacement of antlers, we must consider which level of growth prediction would be most effective.
In the context of predicting changes in deer antler growth, particularly the growth and replacement of antlers, the organismal level (Bone morphology) appears to be the most relevant and effective level of growth prediction. This is because antler growth and replacement involve the development and remodeling of bone structures, which are directly related to the morphology of bones at the organismal level. Therefore, employing techniques used for predicting bone growth and development could potentially provide valuable insights into the changes in deer antler growth, including antler growth and replacement. These methods could be tailored to detect and quantify the changes in antler growth, taking into account the factors such as size, shape, and branching patterns. By integrating these approaches into a predictive model, it may be possible to accurately forecast changes in deer antler growth, thus enhancing our understanding of their growth patterns and contributing to better management and conservation strategies.
Through this SLR, we have deepened our consideration of the potential applicability of GANs for predicting changes in deer antler, such as antler growth and replacement. It should be noted here that this study is not based on cytology to predict growth but is rooted in the CV field, where analysis is performed from images captured by an RGB camera. The composition of deer antlers is divided into layers such as trabecular bone, cortical bone, and mesenchyme. While incorporating the growth characteristics of these compositions into a growth prediction model as physical equations is a possible future direction, it may be insufficient to account for the complex factors affecting bone growth in the wild, such as collisions, injuries, habitat, and food intake. In this regard, machine learning techniques that predict growth stages from images may be more effective in handling these hard cases, as they can learn to recognize and adapt to a wide range of real-world conditions and variations.
GANs, which learns from real-world hard data, has been successfully employed for growth prediction at the cellular level (Tumor cellularity) [ 33, 37] and Tissue and organismal level (Plant growth) [47]. Given this success, it is worth considering the potential applicability of GANs for predicting changes in deer antler, such as antler growth and replacement. In the context of growth prediction, we conclude that GANs has potential to generate synthetic images representing different growth stages, making them a potential candidate for predicting changes in deer antler.
To assess the applicability of GANs for predicting changes in deer antler growth, several factors should be considered: Data availability: A large dataset of deer images at different growth stages, particularly focusing on antler development, is required for training the GANs. This dataset should include a diverse range of images that capture variations in antler size, shape, and branching patterns, as well as other relevant factors that may influence antler growth.
Labeling the antler growth process in stages based on the number of branches and the date of antler growth makes it easy to adapt stage prediction combined with STN, as in the case of [35, 59] et al. First, we will record the individual identification of deer by photographing deer kept in zoos and research facilities. Once the model is trained using these labeled images, we will then apply the learned knowledge to predict antler growth in wild deer populations.
Model adaptation: The GANs architecture and training process should be adapted to the specific task of predicting changes in deer antler. This may involve fine-tuning the model architecture, loss functions, and optimization strategies to ensure that the generated images accurately represent the growth and replacement of antlers. We have not fully elaborated on the details of this process, and it will be the biggest challenge in future research.
Evaluation metrics: Establishing appropriate evaluation metrics is crucial for assessing the performance of the GANs in predicting deer antler changes. These metrics should quantify the similarity between the generated images and the ground truth, taking into account factors such as antler shape, size, and branching patterns. In particular, it is important to comprehensively quantify the size and spacing of branches and the number of branches from size-normalized images, as described below for the pattern of bone branching. This is a reasonable idea since in some papers, the loss function is composed of a composite of indices. For example, in plants, the loss function for color and shape is defined and composited to obtain the final loss function[36]. It is easy to combine the loss functions based on the number of branches, size, color, and other indicators and define a new loss function; however, it is effective according to papers selected by this SLR.
In conclusion, while we conclude that GANs has potential in growth prediction tasks at the Cellular and Tissue and organismal levels, further research and experimentation are needed to determine their applicability for predicting changes in deer antler. By considering the factors mentioned above and adapting the GANs accordingly, it may be possible to develop a successful model for predicting changes in deer antler, including antler growth and replacement.
In addition to GAN-based methods, we also considered the prediction method extracting landmarks from bone images as in the paper by Mahsa Tajdari et al [66]. By specifying the corner points of each vertebra in the 2D frontal (AP) and lateral (LAT) X-ray images, Landmarks are detected with a semi-automatic method. The paper also extracts landmarks from the X-ray images to track non-uniform bone growth. Landmarks are selected for each vertebra, with four corners corresponding to the top and bottom of the growth plate, to obtain a good representation of the bone growth area. They constructed neural networks to predict the growth of cervical spine curvature, using landmark coordinates, global angles representing the overall spine curvature, and patient age as input data.
The above method cannot be directly applied to landmark extraction for deer antlers.
However, it may be possible to establish a landmark representation of the mesenchyme and branches of deer antlers. In that case, we could use a similar neural network that takes landmarks, antler branches, their angles, their total angles, and age as inputs to predict antler shape. Because the method applied in [66] and our method share a similarity in that growth prediction which can be determined by landmarks and angles. Since the method [66] involves manually adjusting automatically generated landmarks at the final stage, manual landmark setting is worth considering in the early stages of research on antler growth prediction. However, antler growth prediction involves a large amount of image data, and it is not practical to manually set landmarks from all the image data. Therefore, we have planned to devise a (semi-)automatic landmark generation technique.
We believe that the revised manuscript conveys this information more effectively and we thank you again for your valuable feedback, which has helped us improve the quality of our manuscript.
Round 2
Reviewer 1 Report
Authors have properly addressed reviewer's comments.
Author Response
Authors have properly addressed reviewer's comments.
Thank you for your positive feedback and acknowledgment that we have adequately addressed your comments. We appreciate your valuable input, which has helped us improve our manuscript.
Reviewer 3 Report
The authors have addressed this reviewer's comments.
