Source: UNIV OF IDAHO submitted to NRP
POTATO WOUND HEALING ENHANCEMENT AND NON-DESTRUCTIVE DETERMINATION BY MEANS OF HYPERSPECTRAL IMAGING (HSI)
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
ACTIVE
Funding Source
AFRI COMPETITIVE GRANT
Reporting Frequency
Annual
Accession No.
1032113
Grant No.
2024-67018-42677
Cumulative Award Amt.
$299,296.00
Proposal No.
2023-10542
Multistate No.
(N/A)
Project Start Date
Sep 1, 2024
Project End Date
Aug 31, 2026
Grant Year
2024
Program Code
[A1364]- Novel Foods and Innovative Manufacturing Technologies
Recipient Organization
UNIV OF IDAHO
875 PERIMETER DRIVE
MOSCOW,ID 83844-9803
Performing Department
(N/A)
Non Technical Summary
Potatoes need to heal their wounds quickly to avoid water loss, diseases, and defects. This study is checking if we can make this happen by healing the wounds at lower curing temperatures. If it works, it can keep the potatoes fresh for longer, which could reduce losses during long-term storage. In the short term, we'll try to heal the wounds in russet potatoes by using calcium chloride and nitric oxide (NO) treatments at lower curing temperatures. We hope this will make the healing process more effective. In the medium term, we'll use a technique called Near-Infrared Hyperspectral Imaging (NIR-HSI) to see how much the wounds have healed. We'll take images of russet potatoes and analyze them to see how much they've healed. We'll use this to create models to predict the healing process. In the long term, we'll use this approach to see if it works in practical settings. We'll evaluate its impact on potato quality, shelf-life, and losses during long-term storage. If it works, it could reduce water loss and decay during prolonged storage. Also, we'll use the NIR-HSI technique to improve storage management practices. It could help us control the environment better and keep the potatoes fresh for longer. Ultimately, these achievements enhance profitability by upholding potato quality, extending shelf life, and curbing losses.
Animal Health Component
30%
Research Effort Categories
Basic
70%
Applied
30%
Developmental
(N/A)
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
5121310102090%
4041310208010%
Knowledge Area
404 - Instrumentation and Control Systems; 512 - Quality Maintenance in Storing and Marketing Non-Food Products;

Subject Of Investigation
1310 - Potato;

Field Of Science
2080 - Mathematics and computer sciences; 1020 - Physiology;
Goals / Objectives
The overall hypothesis of the project is if the wound healing process can be enhanced at low curing temperatures, then potato shelf-life can be extended, and losses minimized during long-term storage. Therefore, the short-term goal of the project is to enhance the wound healing process of Russet potatoes using calcium and nitric oxide at low curing temperatures. The median-term goal is to acquire HSI of Russet potatoes (x, y) and the lignin /suberin content of these samples (λ) and develop prediction models using different chemometric approaches. Finally, the long-term goal is to apply this approach in real-life situations and verify the effect on quality, shelf-life, and potato losses during long-term storage.The specific objectives of this project are:Objective I: (Research) to develop technologies to enhance the wound healing process under low curing temperatures.Objective II: (Research) to develop a non-destructive model to determine the wound healing process by means of HSI and suberin/lignin content in the periderm.By accomplishing these specific objectives, we expect that the rapid wound healing process at low curing temperatures will reduce weight loss and decay in potatoes during long-term storage. We also expect that a non-destructive wound healing measurement using HSI will contribute to the potato industry by improving (on-time / in situ) control processes of the storage environment. All in all, this can increase profitability by maintaining potato quality, extending shelf-life, and reducing losses.
Project Methods
Plant material'Clearwater Russet' potatoes will be used as it is a dual-purpose potato cultivar with good cold-induced sweetening resistance and exhibits excellent fry color out of long-term storage. This cultivar was accepted by McDonald's in 2016 and the acreage has increased tenfold since then.Obj. I: Development of technologies to improve the wound healing process under low curing temperaturesThe general objective of this study is to understand the effect of different products on the biochemistry and physiology of the wound-healing process of potatoes. Thus, an efficient and economical method can be developed to enhance wound healing at low curing temperature conditions and to maintain potato quality during storage.Wound healing experiment: Tubers will be treated with i. control (without any treatment), ii. nitric oxide gaseous treatment (1 mMol/L) for 5 hours in a sealed cabinet with an internal ventilation fan, as suggested by Zhu et al. (2009), and iii. low-volume spray application of 0.2 M solution of food-grade calcium chloride (CaCL2) at a rate of 0.5 gal/ton (Miller et al., 2011). After that, the wound healing process will take place at 10°C / 50°F and 13°C / 55°F with 95% relative humidity (RH) for 14 days. The experiments will be set according to a complete randomized design (CRD) in a factorial arrangement 3 (treatments) x 2 (curing temperatures) x 8 (withdraws, 0, 2, 4, 6, 8, 10, 12, and 14 days) x 3 repetitions of 10 potatoes.Storage experiment: Following the wound healing process, the temperatures will be ramped down to 7.2°C / 45°F, 0.5°F per day. Tubers will be stored at 7.2°C / 45°F and 95% RH for up to 8 months. The experiments will be set according to a complete randomized design (CRD) in a factorial arrangement 3 (treatments) x 2 (curing temperatures) x 5 (withdraws, 0, 2, 4, 6, 8 months) x 3 repetitions of 10 potatoes.Evaluations:a) Wound healing experiment: Fresh weight loss will be determined during the wound healing process every 2 days and on the individual potato tuber cores, this would provide an indirect indication of the impact of wound healing. Wound healing will be evaluated according to the suberin deposition as described by Rui et al. (2021). Periderm samples will be taken every 2 days, immediately frozen using liquid nitrogen, and stored at -80°C / -112°F. After that, periderm samples will be freeze-dried, and ground for the phenylalanine ammonia-lyase (PAL) activity determination following the method described by Flores et al. (2014). Phenolic compounds content will be measured using High-Performance Liquid Chromatography (HPLC) according to Whitehead and Poveda (2019). Fresh tubers will be used for the respiration rate determination (mg CO2 kg-1 h-1) using a static method as described by Saltveit (2003). The RNA-Seq will be conducted at the University of Florida ICBR RNA-sequencing facilities to evaluate the differentially expressed genes (Dr. Liu - collaboration).b) Storage experiment: During the storage period, potatoes will be evaluated every two months for weight loss, specific gravity, sugar content, fry color, and sprouting (Wang et a., 2016).Statistical evaluation: The data will be submitted for analysis of variance and the means will be compared using Tukey's test at a 0.05% probability level using the software R (R Core Team, 2020, Auckland, New Zealand.Obj. II: Development of non-destructive models to determine the wound healing process by means of hyperspectral imaging and suberin/lignin content in the skinThe general objective of this study is to develop a non-destructive model to determine the wound healing process by means of HSI and suberin/lignin content in the periderm. Therefore, a non-destructive wound healing measurement using HSI will contribute to the potato industry by improving (on-time / in situ) control processes of the storage environment.Wound healing experiments (2024-2025 and 2025-2026): Tubers samples from the two-year experiments will be used for the NIR-HIS acquisition. Thus, images of 10 tubers from the previously described treatments (control - without any treatment, 1 mMol/L NO for 5 hours, and 0.2 M CaCL2) will be obtained at days 0, 2, 4, 6, 8, 10, 12, and 14 after the treatment applications.HSI acquisition: The images will be acquired with a Specim FX17 camera (Specim, Spectral Imaging Ltd, Oulu, Finland). The camera comprises an imaging spectrograph coupled to an indium gallium arsenide (InGaAs) detector. Individual images will be collected within a spectral range of 900 to 1700 nm at 8 nm, with spectral sampling per pixel of 3.5 nm. Images of the entire potatoes will be collected at 0, 2, 4, 6, 8, 10, 12, and 14 days after the wound healing treatment applications. Four images will be taken at 8-time intervals. White and dark references will be captured prior to each sample image and subsequently used for image correction and calibration.HSI analysis: The Unscrambler version 10.3 (Camo, Oslo, Norway) will be used for HSI handling and processing. Depending on the case, the hyperspectral data will be processed using Evince v.2.32.0 (UmBio AB, Umeå, Sweden) and MATLAB® 8.9 (The MathWorks Inc., Natick, MA, USA) ambient with PLS Toolbox 8.9 (Eigenvector Research, Inc, Manson, WA, USA) platform and algorithms developed in the laboratory. The image calibration and correction to absorbance will be conducted using Evince package as proposed by Williams et al. (2012). Each individual image will be merged to constitute a mosaic of images. Thus, each mosaic will include the control and the images taken at different time periods for the corresponding treatments.PCA: The regions of interest will be located and identified on the potato skin by applying the PCA score plots, score images, and loading line plots. However, pre-processing techniques such as brushing (Manley et al., 2011) will be previously used to remove irrelevant pixels and recalculate the PCA.PLS: The NIR-HSI of potato skin at different moments of the wound healing process (X matrix) and the suberin/lignin contents (Y matrix) data will be used for the development of the classification and prediction models. As the NIR-HSI of fresh vegetables are highly convoluted and are affected by scattering effects, tissue heterogeneities, instrumental noise, ambient effects, and other sources of variability8, the spectra will be submitted to different pre-processing procedures: standard normal variate, (SNV) and multiplicative scatter correction (MSC) aiming to reduce the influence of light scattering57. Pre-processing can be used isolated or in combination depending on the modeling performance results. The HSI (n=1,920) will be divided into calibration/training (n=1,278) and validation sets (n=641) by applying the classic Kennard-Stone (KS) selection algorithm59. Outliers will also be detected and executed to improve the model's accuracy by removing samples with extreme values, which exhibit increased influence on the model and eliminate unmodelled residues in the X and Y data responses. An elliptical joint confidence region (EJCR) will be calculated to evaluate the slope and intercept for the reference regression and predict values at a 95% confidence interval60,61. PLS with leave-one-out cross-validation will be used and the optimal number of latent variables (LV) will be determined by minimizing the predicted residual error sum of squares (PRESS). The performance of the PLS calibration models will be evaluated using the coefficient of correlation (R2), RMSEC, and cross-validation (RMSECV). With the validation set the RMSEP (SEP) will be obtained8. The images of each potato will be divided into quadrants (rectangles consisting of approximately the same number of pixels), averaged, pre-processed, and the PLS model will be used to obtain the highest number of objects using leave-one-out cross-validation (Williams et al., 2012).

Progress 09/01/24 to 08/31/25

Outputs
Target Audience:The target audience of this project are the professionals in the potato industry which are involved in storage management. This includes growers, agronomists, quality control personnel, shippers, and many others. In addition, this project can benefit professors, researchers, extension specialists and educators, undergraduate and graduate students. Finally, as the project relates to stablishing temperature and treatments to enhancewound healing, professionals in the refrigeration, ventilation, and application industry havealso been targeted. Changes/Problems:We faced problems mostly withdelivery delays. Thus, the camera arrived on September 16th,2024, and we could acquire the NIR-HSI images starting on day two of the wound healing experiment. In addition, the purchase process to acquire the softwareBreeze and Evince from Prediktera AB, Umeå, Sweden, was lengthy and time-consuming, as the Office of Information Technology (OIT) of the University of Idaho needed to approve the purchase and verify for cybersecurity issues. The other main problem was related to the analytical part of the project. As every single individual tuber has been used for the phenolic profile determination, a total of 576 samples in duplicate (n=1,152) need to be evaluated. The PhD student who was hired to develop this project, Hailey Hampton, has extracted and determined all phenolic profile samples, and the last HPLC analysis will be finalized by the end of May 2025. However, all PAL enzymatic activities still need to be carried out. As there is no scientific reason to correlate PAL activity to NIR-HSI imaging, the samples will be pooled together per repetition, and a total of 144 samples in duplicate (n=288) will be used for this evaluation. Finally, the other main problem was related to the RNA-seq analysis. Initial tests made by Dr. Tie Liu's team at the University of Florida resulted in RNA extraction, but the RNA was degraded. Therefore, he was able to modify the method, and a successful extraction was achieved, and we could provide the preliminary results in this report. What opportunities for training and professional development has the project provided?To process and manipulate the hyperspectral data we are using Breeze and Evince packages. In this regard, Prediktera provided three hours training sections, which was attended by one research assistant, one research associate, a M.S. and a PhD student, and one PI. How have the results been disseminated to communities of interest?Yes, the results have been disseminated to the potato industry in many venues, such as: Northwest Potato Research Consortium, Pasco, Washington. 12thFebruary 2025. 57thAnnual Idaho Potato Conference and Eastern Idaho Ag Expo, 2025, Pocatello, Idaho. 23rdJanuary 2025. U of I Potato Storage and Agronomy Advisory Committee Meeting, 2024, Twin Falls, Idaho.03rdDecember 2024. What do you plan to do during the next reporting period to accomplish the goals?Yes, for the next reporting period all goals will be accomplished. Obj. 1: to enhance the wound healing process under low curing temperatures Regarding this objective, we are finalizing the determination of the activity of PAL, total phenolic content and the phenolic compound profile evaluations of all individual potatoes (n=576). This experiment will be repeated in the FY2025-2026 to add more information to the development of the non-destructive wound healing determination. Obj. 2: to determine the wound healing process by means of HSI and suberin/lignin content in the periderm The total phenolic content and phenolic profile (suberin/lignin) will be usedto develop a non-destructive model to determine the wound healing process using NIR-HIS. Classification models and quantification models will be developed according to the proposed methods. To improve the robustness of the prediction models, a second year of data will be incorporated into the dataset.

Impacts
What was accomplished under these goals? Obj. 1: to enhance the wound healing process under low curing temperatures Immediately after the application of the treatments, potatoes were moved to two storage bins set at 50°F or 55°F with 95% relative humidity (RH), for 14 days (Figure 1). Figure 1.Temperatures (°F) of Clearwater Russet tubers cured at 55°F (gray line) or 50°F (orange line), 95% RH for 14 days. a) Wound healing metabolism: During curing (14 days), the weight loss increased for all curing conditions. Curing temperatures did not affect the weight loss of the tubers, but the treatments did (Table 1). Table 1.Weight loss (%) of Clearwater Russet tubers cured at 55°F/96.3% RH or 50°F/97.4% RH, for 14 days. Main factors Weight loss (%) Curing Temperature (A) 50°F 1.05 55°F 0.95 F value 0.3579 Treatments (B) UTC 1.09 a Nitric oxide 0.93 a CaCl2 0.89 ab F value 0.0098 Time - day (C) 0 0.00d 2 0.57c 5 0.85c 9 1.18b 12 1.48 ab 15 1.78a F value <0.0001 Interactions A x B <0.0001 A x C NS B x C NS A x B x C NS CV (%) 33.24 Potatoes from the control treatment showed higher weight loss, but only at 50°F, which might be an indication that the treatments with NO and CaCl2influenced the wound healing process at a lower curing temperature. A fast wound healing process has been related to reduced weight loss in potatoes (Pringles, 2009). Wound healing metabolism needs oxygen and energy from the respiratory process. The respiration rate was not affected by curing temperatures and treatments when expressed as CO2production (Table 2). Table 2.Respiration rate (mg CO2kg-1h-1) of Clearwater Russet tubers at 55°F/96.3% RH or 50°F/97.4% RH, for 14 days. Main factors Respiration rate (mg CO2kg-1h-1) Respiration rate (mg O2kg-1h-1) Curing Temperature (A) 50°F 3.82 5.62 a 55°F 3.83 4.66 b F value 0.9434 <0.0001 Treatments (B) UTC 3.90 5.28 Nitric oxide 3.74 4.97 CaCl2 3.84 5.16 F value 0.3109 0.4786 Time - day (C) 1 4.73 a 7.59 a 4 4.24 b 5.90 b 10 3.27 c 3.94 c 14 3.06 c 3.11 c F value <0.0001 <0.0001 Interactions A x B NS NS A x C 0.0042 0.0078 B x C NS NS A x B x C NS NS CV (%) 44.68 17.15 The lack of differences between the curing temperatures might be related to the stresses inflicted on the tubers during harvest (Figure 1). The determination of the phenylalanine ammonia-lyase (PAL), total phenolic content and the phenolic compound profile are still underway due to the number of samples (n=576) in duplicate (n=1,152). RNA-seq: This analysis was carried out at the University of Florida ICBR RNA-sequencing facilities by Dr. Liu (-Co-PI). The pairwise heatmap of gene ontology (GO) analysis can be seen in Figure 2. Figure 2.Clustered heatmap of gene ontology (GO) enrichment analysis. A volcano plot highlighting genes that meet both LFC threshold of log2fold change > 2 and padj cutoffs of <0.05 can be seen in Figure 3B. Figure 3.Principal component analysis (PCA) of periderm samples from the control at 50°F and the control at 55°F (A). Volcano plot highlighting genes that meet both LFC threshold of log2fold change > 2 and padj cutoffs of <0.05 (B). This indicates that the curing temperature affected the level of gene expression. This analysis is still underway. b) Storage responses: Weight loss: Low curing temperature (50°F/97.4%) resulted in less weight loss than at 55°F/96.3% during storage at 45°F and 95% RH. CaCl2spray application resulted in higher weight loss in relation to UTC. Weight loss increased and reached 5.49% on day 214. Respiration: Potatoes cured at 55°F/96.3% RH had a lower respiration rate during storage (1.49 mg CO2kg-1h-1) than potatoes cured at 50°F/97.4% RH (1.62 mg CO2kg-1h-1), this might be related to the sprout development. Processing quality: One of the main purposes of this project was to answer the question of the impact of low curing temperature (50°F), rather than 55°F, on the processing quality of potatoes. Sprouting: Potatoes cured at 50°F/97.4% RH showed higher sprout development than at 55°F/96.3% RH. The treatments did not affect sprout growth, and sprouts started to develop after 65 days in storage at 45°F. Sugar content: Sucrose content was not affected by the curing temperatures and treatments, but it was affected by the storage period. During storage, a significant reduction in sucrose content was observed, from 0.1037% on day 0 to 0.0383%on days 150 and 210. Tubers cured at 50°F/97.4% RH showed higher glucose content (0.0306%) than at 55°F/96.3% RH (0.0218%). Even though the glucose contents in potatoes cured at 50°F/97.4% RH were higher, these contents were always below the 0.10% threshold value, indicating good processing quality. The reflectance was affected by curing temperatures, but the recorded values were above 41.2, which is considered aUSDA fry color rating of 0. All in all, the quality parameters of Clearwater Russet potatoes cured at 50°F/97.4% RH or 55°F/96.3% RH at 45°F, 95% RH for 210 days, were comparable with previous studies. Obj. 2: to determine the wound healing process by means of HSI and suberin/lignin content in the periderm As soon as we received the proposal approval, we took all necessary actions to quickly purchase theSpecim FX17 camera, the LabScanner with dual illumination package, Breeze® and Evince® software. Unfortunately, due to delivery delays, the camera arrived on September 16th,2024, and we could acquire the NIR-HSI images starting on day two of the wound healing experiment (Figure 4). Figure 4.Specim FX17 camera and LabScanner with dual illumination package. Quantification models: As the total phenolic content and phenolic profile (suberin/lignin) are still under analysis, the dry matter (DM) content was used to develop a non-destructive model to determine the wound healing process using NIR-HIS (Figure 5). During curing, the accumulation of phenolic compounds in the periderm might be related to the increase in DM content (Campilho et al., 2020). Dry matter content increased significantly from 17.2% on day 0 to 20.0% on day 12, with no significant difference between days 12 (20.0%) and 14 (20.2%) irrespective of treatments and curing temperatures. This might indicate that wound healing was completed between days 12 and 14. Partial least square regression (PLS) using full cross-validation resulted in a root mean square error of calibration (RMSECV) of 0.008%, a coefficient of determination for calibration (Rc2) of 0.73, and a root mean square error of prediction (RMSEP) of 0.0081%, and a coefficient of determination for prediction (Rp2) of 0.70 (Figure 5C). ? Figure 5. Non-destructive classification of Clearwater Russet potato tubers based on curing temperatures (10°C/50°F or 12.7°C/55°F, 95% RH for 14 days) and dry matter (DM) prediction using NIR-HSI images. (A) NIR-HSI image, (B) NIR-HSI processed image, and (C) PLS dry matter content prediction. These results demonstrate that NIR-HSI offers a non-destructive method for determining DM content in intact potato tubers, which might be a valid tool for monitoring the wound healing process. A similar approach will be used for the total phenolic content and phenolic profile (suberin/lignin) prediction models.

Publications

  • Type: Conference Papers and Presentations Status: Published Year Published: 2024 Citation: Pedrosa, V.M.D., Izidoro, M., Liu, T., Paytosh, S., Olsen, N., Teixeira, G.H.A. (2024). Curing temperatures affect the wound healing metabolism of Clearwater Russet potato tubers. IX International Postharvest Symposium, Rotorua, New Zealand, p.115. https://airdrive.eventsair.com/eventsairaueprod/production-scienceevents-public/1791dd556e934c20b9a8fee5b2a9558b