At Oceanic Laboratory Initiative (OLI), we embrace the power of Artificial Intelligence (AI) to revolutionise mould analysis. Our AI analysis is meticulously validated and edited by qualified analysts before any report is issued, ensuring accuracy and reliability.
Unlike traditional methods, AI does not conform to recognised standards for analysis, as these standards are tailored for human processes, manually operated equipment, and personal analyst capacities. AI operates beyond these constraints, offering a more dynamic and efficient approach.
In AI-based mould analysis, achieving accurate results hinges on striking a delicate balance between two crucial metrics: precision and recall. These two metrics play a fundamental role in ensuring that the AI model identifies and classifies mould structures with the highest level of accuracy.
Precision:
Precision in AI analysis refers to the ability of the model to correctly identify mould structures when it detects them. In other words, it measures how many of the identified instances are truly relevant. A high precision score means that when the AI identifies mould structures, it is highly likely to be accurate, minimizing false positives. This is especially critical in mould analysis, as false positives can lead to unnecessary concerns and actions.
Recall:
Recall, on the other hand, is the ability of the AI model to find all the relevant mould structures in a given sample. It measures how well the model detects and captures all instances of mould, including those that might be less obvious or concealed. High recall ensures that no significant mould structures are missed during analysis, preventing false negatives where mould might be present but goes undetected.
Achieving a perfect balance between precision and recall can be challenging because they are often inversely related. Improving precision may lead to a decrease in recall and vice versa. Finding the right equilibrium is crucial, as it ensures that the AI model accurately identifies mould while minimising both false positives and false negatives.
The effectiveness of precision and recall in AI mould analysis depends on the quality of the data used to train the model. High-quality, well-labeled training data contribute to a model that can discern mould structures from non-mould elements with precision and can recognize mould structures across various scenarios with high recall..
At Oceanic Laboratory Initiative (OLI), our AI model is continuously learning and improving. By training on a vast dataset that includes diverse mould presentations, it refines its ability to balance
precision and recall. As the model encounters new data, it adapts and fine-tunes its performance, striving for optimal accuracy in mould analysis.
This ongoing learning process ensures that the AI-based mould analysis offers the highest levels of precision and recall, providing clients with reliable and comprehensive results, even in complex indoor environments.
Our AI model has been trained on an extensive dataset of over 200,000 sample images, with ongoing growth. This continuous learning process allows the model to consistently improve, enabling more comprehensive outputs as the database expands. This enhancement extends beyond mould identification and benefits other aspects of Indoor Air Quality analysis.
In summary, precision and recall are key metrics in AI mould analysis, representing the model’s ability to accurately identify mould structures while minimising false positives and false negatives. Striking the right balance and continuously improving this balance is essential for delivering trustworthy mould analysis results.
Our commitment to quality includes a validation and editing process of the software outputs and is performed by our team of highly qualified analysts. Before any report is issued, our experts carefully review and fine-tune the results generated by AI, ensuring the utmost precision and reliability in our findings getting some of the best of both worlds in the final report.
It’s worth noting that AI analysis, while powerful, does not adhere to conventional recognised standards for analysis like the ASTM D7391 – Standard Test Method for Categorisation and Quantification of Airborne Fungal Structures in an Inertial Impaction Sample by Optical Microscopy. This type of standard was developed for human processes, manually operated equipment, and personal analyst capacities, clarifying the standard operating procedure for this type of analysis. AI does not conform to these constraints, yet still offering a level of analysis that would be that has defensibility with our team proofing every sample.
For air sample analysis, we currently support Allergenco-D and Air-O-Cell cassettes. These cassettes are compatible with our AI-based analysis method, ensuring accurate results.
We believe in harnessing the potential of AI to augment our analysis capabilities. Just like a child grows into a smarter adult, AI learns and improves over time. Unlike regular programming, AI adapts to new scenarios, making it more versatile and efficient.
AI’s strength lies in its ability to process vast amounts of data, identifying patterns, making predictions, and recommending actions, much like a human but faster and more precise. At OLI, we’ve collected a massive dataset of tagged images, feeding it to our AI software to enhance its mould detection capabilities.
Traditional analysis methods may miss spore clusters, leading to variations in counts. However, the AI technology analyses 100% of every sample, eliminating operator bias and ensuring consistent, deterministic results regardless of when the sample is tested.
At Oceanic Laboratory Initiative (OLI), we are committed to delivering reliable non-culturable mould analysis services that meet industry standards and exceed expectations.
We’re here to assist you every step of the way. Whether you have questions or need expert insights, our team is just a call or message away.
Mould Testing Analysis Servicing the Oceanic Region
For purchasing enquiries please email: lab@olilabs.com.au or fill out a contact form