Machine Learning to Detect Fungal Infections in Stored Pome Fruits via Mass Spectrometry Data - Industry, Economic, and Social Implications

Aug 5, 2026

Upload by Violet

Pome fruits such as apples and pears can decay during storage due to fungal infections, making early and reliable detection essential to protect fruit quality in warehouses. An experimental study inoculates two apple cultivars with Monilinia laxa, Neonectria ditissima, and Botrytis cinerea, then captures infection-related volatile chemicals using mass spectrometry in positive and negative ion modes. Machine learning classifiers using multiple features across Apple-Fungi-Ion combinations are evaluated by accuracy and error types, showing Logistic Regression and linear-kernel SVM as top performers in most cases. Results indicate high practical value for industrial safeguarding, economic loss reduction, and safer market timing for consumption.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages10

Reading Time25 min

AccessFree to download

Abstract

I. Introduction

Machine Learning to Detect Fungal Infections in Stored Pome Fruits via Mass Spectrometry Data: Industry, Economic, and Social Implications Razia Sulthana Abdul Kareem 1, *, Nageena K. Frost 1, Charles A. I. Goodall 2, Timothy Tilford 1, and Ana Paula Palacios 1 1 School of Computing and Mathematical Sciences, Faculty of Engineering and Science, University of Greenwich, Old Royal Naval College, London, United Kingdom 2 Faculty of Engineering and Science, University of Greenwich, Chatham Maritime, Chatham, United Kingdom Email: [razia.sulthana@greenwich.ac.uk](razia.sulthana@greenwich.ac.uk) (R.S.A.K.); [n.k.frost@greenwich.ac.uk](n.k.frost@greenwich.ac.uk) (N.K.F.); [c.a.i.goodall@greenwich.ac.uk](c.a.i.goodall@greenwich.ac.uk) (C.A.I.G.); [t.tilford@greenwich.ac.uk](t.tilford@greenwich.ac.uk) (T.T.); [a.palacios@greenwich.ac.uk](a.palacios@greenwich.ac.uk) (A.P.P.) *Corresponding author Abstract—Pome fruits, notably apples and pears, experience decay during storage due to fungal infections. The timely discernment of these infections is imperative to avert the deterioration of these fruits within warehouse confines. In an experimental setup, two distinct apple cultivars, Braeburn and Gala, were inoculated with fungi Monilinia laxa, Neonectria ditissima, and Botrytis cinerea. As the infection progresses, the apples release chemical volatile components, which are measured using mass spectrometry in both positive and negative ion modes, recording mass-charge ratios ranging from m/z 30 to m/z 900 with a 0.3 Dalton difference between each measurement. The dataset is then partitioned into 24 sets of three-dimensional data, encompassing attributes related to two types of apples, three types of fungi, and two types of ions. They are analyzed using various machine learning algorithms, including Logistic Regression, Support Vector Machines (SVM), XGBoost, Random Forest, and four distinct customised Neural Networks, to classify infected and uninfected apples. The outcomes from the different machine learning algorithms across the 12 combinations of Apple-Fungi-Ion are recorded, revealing that certain algorithms excel in different combinations. The performance metrics namely True Positive, True Negative, False Positive, False Negative, Accuracy are closely analysed and the algorithms that produce the highest and secondhighest accuracy are highlighted. Upon thorough analysis of the 12 combinations, it is observed that Logistic Regression and SVM with a linear kernel achieve the highest accuracy in approximately 11 combinations. Specifically, Logistic Regression achieves a precision of 98% for Braeburn apples, while SVM attains 99% accuracy for Gala apples. This research project has a triple impact on industry, economy, and society. On an industrial level, the precision and early predictions of the proposed work can effectively safeguard large quantities of apples in storage bins. Economically, it has the potential to avert substantial monetary losses. Societally, it plays a crucial role in determining the ideal timing to release fruits to the market for consumption without jeopardizing human health. Manuscript received June 20, 2024; revised July 2, 2024; accepted July 30, 2024; published October 23, 2024. Keywords—pome fruits, apples, fungal infection, mass spectrometry, machine learning algorithms, neural networks I. INTRODUCTION Among the plethora of available fruits, apples emerge as the preeminent choice for individuals spanning all age groups, proving beneficial to intestinal health amidst various illnesses. Apples are convenient for on-the-go consumption at work or during travel and the judicious interplay between the price and the size of apples renders them reasonably priced and affordable [1] . Numerous countries globally have implemented legal regulations governing the cultivation, processing, packaging, and fresh delivery of diverse apple cultivars. In 2011, the United Kingdom established a standard to deliv

FAQs

How are fungal infections detected in stored pome fruits in this study?

Infected and uninfected apples are distinguished using volatile chemical components measured by mass spectrometry, followed by machine learning classification based on the resulting spectral data.

Which machine learning models are compared for classifying infected vs. uninfected apples?

The study compares Logistic Regression, Support Vector Machines (SVM), XGBoost, Random Forest, and multiple customised Neural Networks across different Apple-Fungi-Ion data combinations.

What are the main industrial, economic, and social implications described?

Early and accurate predictions can protect large quantities of apples during storage (industrial), prevent substantial monetary losses (economic), and support the right timing for releasing fruits to the market without jeopardizing human health (societal).

Aug 5, 2026

Upload by Violet