KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Search Results

Now showing 1 - 8 of 8

Some of the metrics are blocked by your
consent settings
Explainable machine learning reveals diverse yield-determining factors among Thai rice farmer cohorts: Implications for targeted agricultural support
(2026-06-01)
Suriyalaksh, Manusnan
;
Prommawin, Benjapon
;
Hannanta-Anan, Pimkhuan
;
Sreewongchai, Tanee
;
Promchote, Parichart
Rice yield prediction and optimization remain crucial challenges in Thailand’s agricultural sector. This study presents an explainable machine learning framework for predicting farm-level rice yields and identifying key factors affecting productivity. We collected comprehensive data from 1,722 smallholder farmers in central Thailand, encompassing 58 agronomic and economic variables. Four automated machine learning (AutoML) frameworks – AutoGluon, auto-sklearn, h2o, and mljar – were evaluated using 5-fold cross-validation, with AutoGluon achieving the best performance (root mean square error: 0.532 tonnes/hectare, mean absolute error: 0.372 tonnes/hectare, R<sup>2:</sup> 0.538). Using global SHapley Additive exPlanations (SHAP) analysis, we identified farmers’ social networks, rental costs during harvest, and total harvesting expenses as the most influential predictors of rice yields. Notably, stronger social network connectivity was associated with higher yields, suggesting that information sharing and collective knowledge within farming communities play a key role in improving productivity. Clustering analysis based on individual SHAP values revealed six distinct farmer cohorts, each characterized by unique patterns of feature importance. These cohort-specific insights demonstrate the potential of combining AutoML with explainability techniques to move beyond uniform agricultural recommendations towards precision support tailored to the specific needs of different farmer cohorts.
Some of the metrics are blocked by your
consent settings
A Cost-Effective Digital Microfluidic System for CRISPR Diagnostic Automation
(2024-01-01)
Vunkong, Jirachaya
;
Hannanta-Anan, Pimkhuan
CRISPR diagnostics offer precise detection of genetic material with high specificity and accuracy while adaptable for field deployment. However, the labor-intensive steps required can reduce usability and increase the risk of human error. To overcome these limitations, we developed a low-cost digital microfluidic (DMF) platform integrated with a custom fluorescence imaging module for the automation of CRISPR diagnostics. Our DMF platform was built from a basic PCB array coated with a layer of parafilm and silicone oil. The fluorescence module was custom designed and fabricated through 3D printing. Our integrated platform effectively facilitated movement, mixing, and fluorescence imaging of water and CRISPR-Cas reagents. This affordable DMF system demonstrates its initial feasibility for automating CRISPR diagnostics, enhancing accessibility and user- friendliness.
Some of the metrics are blocked by your
consent settings
Fabrications of Low-Cost Poly(methyl methacrylate) Microfluidic Devices for Precise Passive Flow Control
(2024-01-01)
Kaewtae, Wattana
;
Tanasapsakul, Warinthorn
;
Pattamang, Pattaraluck
;
Ranron, Norabadee
;
Sripumkhai, Witsaroot
Microfluidics serve as effective platforms for conducting point-of-care diagnostics and biochemical assays which require precise control over flow rates and incubation times. To enable their uses in resource-limited settings, a pump-free approach is essential for driving fluid flow within the microfluidics. In this study, we introduced cost-effective methodologies for fabricating poly(methyl methacrylate) (PMMA)-based microfluidic devices driven passively by capillary pressure. We explored three fabrication techniques including two-layer CNC micro-milling, two-layer laser engraving, and three-layer laser cutting. Among these methods, three-layer laser cutting proved to be the most reproducible method, yielding devices with consistent channel dimensions and contact angles. Consequently, this fabrication technique enables better control over flow rates within microfluidic systems. Our findings demonstrate the feasibility of fabricating low-cost, pump-free microfluidic devices for precise flow control, with potential applications in disease detection and biomedical research.
Some of the metrics are blocked by your
consent settings
Computational framework for single-cell spatiotemporal dynamics of optogenetic membrane recruitment
(2022-07-18)
Kuznetsov, Ivan A.
;
Berlew, Erin E.
;
Glantz, Spencer T.
;
Hannanta-Anan, Pimkhuan
;
Chow, Brian Y.
We describe a modular computational framework for analyzing cell-wide spatiotemporal signaling dynamics in single-cell microscopy experiments that accounts for the experiment-specific geometric and diffractive complexities that arise from heterogeneous cell morphologies and optical instrumentation. Inputs are unique cell geometries and protein concentrations derived from confocal stacks and spatiotemporally varying environmental stimuli. After simulating the system with a model of choice, the output is convolved with the microscope point-spread function for direct comparison with the observable image. We experimentally validate this approach in single cells with BcLOV4, an optogenetic membrane recruitment system for versatile control over cell signaling, using a three-dimensional non-linear finite element model with all parameters experimentally derived. The simulations recapitulate observed subcellular and cell-to-cell variability in BcLOV4 signaling, allowing for inter-experimental differences of cellular and instrumentation origins to be elucidated and resolved for improved interpretive robustness. This single-cell approach will enhance optogenetics and spatiotemporally resolved signaling studies.
Some of the metrics are blocked by your
consent settings
Multi-Agent Q-Leaming for Power Allocation in Interference Channel
(2022-01-01)
Wongphatcharatham, Tanutsorn
;
Phakphisut, Watid
;
Wijitpornchai, Thongchai
;
Areeprayoonkij, Poonlarp
;
Jaruvitayakovit, Tanun
Signal transmission in wireless networks suffers from unwanted interference. To maximize signal to interference plus noise ratio, transmit power of each transmitter needs to be optimally allocated. Here, we propose to use multi-agent Q-learning to optimize such transmit power within interference channel. Our simulation indicated that multi-agent Q-Iearning resulted in better sum-rate than the traditional methods such as the maximum power allocation and the random power allocation. Our work offers a novel and practical computational approach to optimizing signal transmission in wireless networks.
Some of the metrics are blocked by your
consent settings
Computer Vision-aided CRISPR Diagnostics for the Detection of COVID-19
(2021-04-01)
Nimsamer, Pattaraporn
;
Mayuramart, Oraphan
;
Samacoits, Aubin
;
Chantaravisoot, Naphat
;
Nakhakes, Chajchawan
Surveillance testing is a key strategy to control the spread of COVID-19. Unlike the gold standard testing method, quantitative reverse transcription polymerase chain reaction (RT-qPCR), CRISPR diagnostics have recently become a more appealing alternative as they are proven to be faster, simpler, and more affordable. However, the current CRISPR diagnostic readouts are typically non-quantitative, making them error-prone and lacking crucial information of viral load. To further improve the CRISPR diagnostic method, we have developed a custom computer vision algorithm that works in complement to common transilluminators to process fluorescence images of the diagnostic samples, quantify their fluorescence signals, and assign the test results. Our analysis showed that the quantified fluorescence intensity was directly correlated to the sample viral load, useful information for transmissibility and disease severity. Verified through laboratory and clinical samples, our algorithm accurately discriminated the samples with the viral RNA as low as 6.25 copies/uL, and correctly classified nasopharyngeal swab (NP swab) samples with 100% accuracy. Our work serves as a potential technique to improve the accuracy of CRISPR diagnostics of COVID-19 and promote rapid testing vital to the containment of the ongoing pandemic.
Some of the metrics are blocked by your
consent settings
Machine Learning-Driven and Smartphone-Based Fluorescence Detection for CRISPR Diagnostic of SARS-CoV-2
(2021-02-02)
Samacoits, Aubin
;
Nimsamer, Pattaraporn
;
Mayuramart, Oraphan
;
Chantaravisoot, Naphat
;
Sitthi-Amorn, Pitchaya
Rapid, accurate, and low-cost detection of SARS-CoV-2 is crucial to contain the transmission of COVID-19. Here, we present a cost-effective smartphone-based device coupled with machine learning-driven software that evaluates the fluorescence signals of the CRISPR diagnostic of SARS-CoV-2. The device consists of a three-dimensional (3D)-printed housing and low-cost optic components that allow excitation of fluorescent reporters and selective transmission of the fluorescence emission to a smartphone. Custom software equipped with a binary classification model has been developed to quantify the acquired fluorescence images and determine the presence of the virus. Our detection system has a limit of detection (LoD) of 6.25 RNA copies/μL on laboratory samples and produces a test accuracy of 95% and sensitivity of 97% on 96 nasopharyngeal swab samples with transmissible viral loads. Our quantitative fluorescence score shows a strong correlation with the quantitative reverse transcription polymerase chain reaction (RT-qPCR) Ct values, offering valuable information of the viral load and, therefore, presenting an important advantage over nonquantitative readouts.
Some of the metrics are blocked by your
consent settings
An ultra-violet sterilization robot for disinfection
(2019-07-01)
Chanprakon, Pacharawan
;
Sae-Oung, Tapparat
;
Treebupachatsakul, Treesukon
;
Hannanta-Anan, Pimkhuan
;
Piyawattanametha, Wibool
Ultraviolet (UV) sterilization technology is used to aid in reduction of microorganisms that may remain on the surfaces after a standard cleaning to the minimum number. Our research team developed a UV robot or UV bot for sterilization in an operating or a patient room. Our UV bot has three 19.3-watt of UV lamps mounted on top of the UV bot platform covering 360° direction. Our UV bot employed an embedded system based on a Raspberry Pi to aid in navigation to avoid obstacles. In addition, we tested the effectiveness of eliminating Staphylococcus Aureus bacteria sample plates located 35 cm away from our UV bot to be within 8 seconds after UV light exposure.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results