Bacterial biofilms on wounds lead to severe infections because of their resistance to antibiotics and host defence mechanisms. These bacterial colonies comprising of bacterial cells embedded in an extracellular matrix can easily develop on wounds, are hard to detect, and significantly delay wound healing leading to chronic infections especially in diabetic foot ulcer patients. Current methods to detect bacterial biofilms are extremely cumbersome and time-consuming (2-3 days) and, therefore, pose a challenge to low-resource implementations. We demonstrate a machine learning aided rapid wound blot detection method (<10 min) that combines the wound blotting technique using nitrocellulose membranes, white light imaging and machine learning-based models to accurately infer the presence of biofilms. We validate our method against the standard test-tube method that utilizes dye staining of the wound-swab culture to infer biofilm presence and demonstrate a detection accuracy in excess of 85%.
Rapid, point-of-care bacterial bioburden detection on diabetic foot ulcers, burn wounds, and surgical site wounds, and subsequent gram-type classification is a challenge that cannot be addressed in low-resource settings. This is because current gold-standard practices based on cell culture take 2-3 days for a pathogen detection and classification. Further, these methods are cumbersome, require sophisticated microbiological facilities, and trained personnel. To address this issue, we demonstrate a rapid (<2 min), point-of-care platform, Illuminate®, that combines autofluorescence intensity imaging and deep learning to detect bacterial bioburden. We use a UNET deep learning model with a DenseNet 201 backbone to detect bioburden spatially and, to further classify the gram-type. Our initial results show a >80% accuracy in detecting bioburden and a >70% accuracy in gram-type classification. We further use this platform for identifying clinically relevant bacteria such as pseudomonas aeruginosa that cause antimicrobial resistance.
Terahertz and autofluorescence imaging technologies are combined for accurate breast and oral cancer margin detection. More than thirty fresh tissue samples are imaged in this study. Cancer progression causes structural, and metabolic changes which can be probed effectively by combining Terahertz and Autofluorescence technologies and using advanced machine learning algorithms. To train the Machine Learning algorithm, the cancer and noncancer regions in Terahertz and fluorescence images are identified by overlapping with histopathology images. This study confirms that the combination of multiple spectroscopy techniques and Machine Learning algorithms has the potential to achieve better diagnostic accuracy in fresh cancer tissue.
Bioburden detection in water is an important challenge in the contexts of both consumer and industrial water. Water-borne infections due to bacteria and fungi are becoming key public health concerns. Further, in biologics manufacturing sector, it is of key importance to use water with zero bioburden in all critical manufacturing processes. However, current methods of detecting and classifying bioburden in water samples is a tedious process involving time-consuming microbiological steps where it takes about 5-7 days to infer trace levels of pathogen contamination. It is possible to hasten the detection process using cytometry-based platforms that offer high levels of sensitivity and specificity in pathogen detection and enumeration of various pathogens. However, these solutions may not be able to determine the bacterial species and viability in a label-free set up. Here, we present Illuminate-τ, a portable label-free detection platform for rapid bioburden assessment in water samples in under 10 minutes, thereby demonstrating a >1000x improvement in detection time. This works on the principle of inferring the presence of pathogen cells in water through their native autofluorescence lifetime characteristics. The Illuminate-τ device is integrated with pulsed nanosecond UV light sources and drivers, high-speed single photon avalanche detectors, and sophisticated timing circuits controlled by an embedded electronics subsystem. In addition, the device also has an edge inferencing SVM-based machine learning algorithm that takes in autofluorescence lifetime characteristics and detects <100 CFU/ml of bioburden such as Pseudomonas, E.coli, Salmonella, Candida species etc. with an accuracy exceeding 99%. Further, we show that we the platform is also able to differentiate Pseudomonas from other bacteria and fungi with a 100% accuracy under similar conditions of concentration. In summary, we demonstrate that Illuminate-τ a device based on autofluorescence lifetime coupled with machine-learning-based detection strategies can achieve high bioburden detection sensitivity.
Pathogens such as bacteria and fungi express various autofluorescence markers such as NAD(P)H, flavins, porphyrins etc. In contrast to steady state fluorescence, wherein the amplitude variations are more sensitive to bacterial growth state, time dependant fluorescence characteristics of these autofluorescence biomarkers are more robust and can potentially provide higher specificity of detection especially in various clinical and industrial settings. However, fluorescence lifetime measurements are typically carried out using bulky and expensive instruments. Herein, we demonstrate an extremely portable device 10 cm*10 cm*5 cm for fluorescence lifetime measurement comprising of pulsed LED sources, optical filters and single photon avalanche photodetectors (SPADs). The entire instrument is driven by an embedded hardware board and the photon arrival times from the SPAD are measured using a Texas instrument TDC counter interfaced with the hardware. First, the instrument response function of the device is measured and is found to be <1 ns suitable for measuring the fluorescence lifetime of various pathogens. Subsequently, fluorescence lifetime measurements of pathogens such as Staphylococcus Aureus, Pseudomonas Aeruginosa and E.coli is done under 365 nm light excitation. Using the autofluorecence lifetime characteristics and classification algorithms, we demonstrate bioburden detection in water with sensitivity exceeding 90% with a minimum detection limit of 103 CFU/mL. Furthermore, a handheld version of the device is also developed for assessment of the pathogens directly on the clinical samples such as ulcers, burns, surgical site infections. Notably, the handheld device is able to track bacterial cell growth, differentiate cancer tissues from noncancerous samples and is also able to detect microbial growth in flesh samples demonstrating its applicability in a range of potential applications.
Anti-microbial resistance has emerged as a major global threat. Due to lack of rapid AST tests, generic antibiotics are typically prescribed. Pathogens possess various metabolic biomarkers such as NAD(P)H and flavins which exhibit auto fluorescence. A rapid phenotypic point-of-care AST device is developed that leverages the changes in the autofluorescence when the pathogens encounter antibiotic stress. The device is integrated with multiple wavelength sources to excite various biomarkers, and a CMOS camera integrated with optical filters to capture the emitted autofluorescence intensity. The results show that the device is capable of determining antibiotic susceptibility with its significant minimum inhibitory concentration in under 5 hours suitable for point-of-care testing.
Skin and soft tissue infections (SSTIs) are one of the most common infections in India affecting 10-12% of Indian population. They are caused by a variety of bacteria and fungus, which makes it harder to diagnose and propose an effective treatment immediately especially in low resource settings due to the lack of access to qualified physicians. Management of SSTIs requires early expert infection assessment and remains a major challenge for the clinicians. A hand-held device is developed leveraging the inherent autofluorescence properties of the bacterial and fungal species that can non-invasively and rapidly identify the pathogens on SSTI using multispectral imaging followed by image processing and machine learning algorithms. The device is able to classify the gram type with < 85% accuracy.
Diabetic foot ulcers are common, recurrent, leading frequently to foot amputation and even death. Their management requires early expert infection assessment and remains a major challenge for the clinicians. Assessment also necessitates culture-sensitivity of the swab taken from ulcer (the gold-standard technique) to identify the bacteria colonizing the infected wound. The process requires accurate swabbing, culturing in a BSL-2 facility and takes anywhere between 2-5 days leading to prescription of generic antibiotics by the doctors. Regular swabbing is a cumbersome procedure to understand and regularly follow up on the microflora population.
Each bacteria has characteristic emission fluorescence when excited with different wavelength of light sources. A novel device, developed by us, leverages this auto-fluorescence property enabling us to develop a multispectral imaging platform. The device captures the spectral signatures of metabolic growth markers along with markers released when a microbiome causes infection to detect and assess the bacterial gram type.
A preliminary clinical study was conducted at MV Hospital for Diabetes and Prof M Viswanathan Diabetes Research Centre, Chennai. Of the 50 patients imaged, the spectral signatures obtained from our device was able to find significant differences between gram positive and gram-negative bacteria. The device spectral results was compared against deep tissue culture biopsy and the device was able to detect gram positives and gram negatives with 83% and 81% accuracy respectively. The device also picked up 7 polymicrobial sites.
In summary, the device can be used as an important tool in guided swabbing, assessment of a wound and understanding its microbiome pattern. The device helps to differentiate infected from non infected wounds, classifies the infected ones broadly according to their gram type and enables real time follow up of wounds. In future, fluorescence spectral signatures will be obtained using more excitation wavelengths to differentiate the exact species of bacteria and to improve on the accuracy of classification to enable treatment protocols using tailored antibiotics.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.