Acute respiratory distress syndrome (ARDS) is the most severe form of acute lung injury, responsible for high mortality and long-term morbidity. As a dynamic syndrome with multiple etiologies, its timely diagnosis is difficult as is tracking the course of the syndrome. Therefore, there is a significant need for early, rapid detection and diagnosis as well as clinical trajectory monitoring of ARDS. Here, we report our work on using human breath to differentiate ARDS and non-ARDS causes of respiratory failure. A fully automated portable 2-dimensional gas chromatography device with high peak capacity (> 200 at the resolution of 1), high sensitivity (sub-ppb), and rapid analysis capability (~30 min) was designed and made in-house for on-site analysis of patients' breath. A total of 85 breath samples from 48 ARDS patients and controls were collected. Ninety-seven elution peaks were separated and detected in 13 min. An algorithm based on machine learning, principal component analysis (PCA), and linear discriminant analysis (LDA) was developed. As compared to the adjudications done by physicians based on the Berlin criteria, our device and algorithm achieved an overall accuracy of 87.1% with 94.1% positive predictive value and 82.4% negative predictive value. The high overall accuracy and high positive predicative value suggest that the breath analysis method can accurately diagnose ARDS. The ability to continuously and non-invasively monitor exhaled breath for early diagnosis, disease trajectory tracking, and outcome prediction monitoring of ARDS may have a significant impact on changing practice and improving patient outcomes. Keywords Breath analysis. Acute respiratory distress syndrome gas chromatography. 2D GC. Machine learning Menglian Zhou and Ruchi Sharma contributed equally to this work.

Enzymatic colorimetric analysis of metabolites provides signatures of energy conversion and biosynthesis associated with disease onsets and progressions. Miniaturized photodetectors based on emerging two-dimensional transition metal dichalcogenides (TMDCs) promise to advance point-of-care diagnosis employing highly sensitive enzymatic colorimetric detection. Reducing diagnosis costs requires a batched multisample assay. The construction of few-layer TMDC photodetector arrays with consistent performance is imperative to realize optical signal detection for a miniature batched multisample enzymatic colorimetric assay. However, few studies have promoted an optical reader with TMDC photodetector arrays for on-chip operation. Here, we constructed 4 × 4 pixel arrays of miniaturized molybdenum disulfide (MoS2) photodetectors and integrated them with microfluidic enzyme reaction chambers to create an optoelectronic biosensor chip device. The fabricated device allowed us to achieve arrayed on-chip enzymatic colorimetric detection of d-lactate, a blood biomarker signifying the bacterial translocation from the intestine, with a limit of detection that is 1000-fold smaller than the clinical baseline, a 10 min assay time, high selectivity, and reasonably small variability across the entire arrays. The enzyme (Ez)/MoS2 optoelectronic biosensor unit consistently detected d-lactate in clinically important biofluids, such as saliva, urine, plasma, and serum of swine and humans with a wide detection range (10–3–103 μg/mL). Furthermore, the biosensor enabled us to show that high serum d-lactate levels are associated with the symptoms of systemic infection and inflammation. The lensless, optical waveguide-free device architecture should readily facilitate development of a monolithically integrated hand-held module for timely, cost-effective diagnosis of metabolic disorders in near-patient settings.

Acute respiratory distress syndrome (ARDS) is the most severe form of acute lung injury, responsible for high mortality and long-term morbidity. As a dynamic syndrome with multiple etiologies its timely diagnosis is difficult as is tracking the course of the syndrome. Therefore, there is a significant need for early, rapid detection and diagnosis as well as clinical trajectory monitoring of ARDS. Here we report our work on using human breath to differentiate ARDS and non-ARDS causes of respiratory failure. A fully automated portable 2-dimensional gas chromatography device with high peak capacity, high sensitivity, and rapid analysis capability was designed and made in-house for on-site analysis of patients' breath. A total of 85 breath samples from 48 ARDS patients and controls were collected. Ninety-seven elution peaks were separated and detected in 13 minutes. An algorithm based on machine learning, principal component analysis (PCA), and linear discriminant analysis (LDA) was developed. As compared to the adjudications done by physicians based on the Berlin criteria, our device and algorithm achieved an overall accuracy of 87.1% with 94.1% positive predictive value and 82.4% negative predictive value. The high overall accuracy and high positive predicative value suggest that the breath analysis method can accurately diagnose ARDS. The ability to continuously and non-invasively monitor exhaled breath for early diagnosis, disease trajectory tracking, and outcome prediction monitoring of ARDS may have a significant impact on changing practice and improving patient outcomes.

Background: In this study, we examined the ability of resonance Raman spectroscopy to measure tissue hemoglobin oxygenation (R-StO 2 ) noninvasively in critically ill patients and compared its performance with conventional central venous hemoglobin oxygen saturation (ScvO 2 ). Methods: Critically ill patients (n ¼ 138) with an indwelling central venous or pulmonary artery catheter in place were consented and recruited. R-StO 2 measurements were obtained by placing a sensor inside the mouth on the buccal mucosa. R-StO 2 was measured continuously for 5 min. Blood samples were drawn from the distal port of the indwelling central venous catheter or proximal port of the pulmonary artery catheter at the end of the test period to measure ScvO 2 using standard co-oximetry analyzer. A regression algorithm was used to calculate the R-StO 2 based on the observed spectra. Results: Mean (SD) of pooled R-StO 2 and ScvO 2 were 64(7.6) % and 65(9.2) % respectively. A paired t test showed no significant difference between R-StO 2 and ScvO 2 with a mean(SD) difference of À1(7.5) % (95% CI: À2.2, 0.3%) with a Clarke Error Grid demonstrating 84.8% of the data residing within the accurate and acceptable grids. Area under the receiver operator curve for R-StO 2 's was 0.8(0.029) (95% CI: 0.7, 0.9 P < 0.0001) at different thresholds of ScvO 2 ( 60%, 65%, and 70%). Clinical adjudication by five clinicians to assess the utility of R-StO 2 and ScvO 2 yielded Fleiss' Kappa agreement of 0.45 (P < 0.00001). Conclusions: R-StO 2 has the potential to predict ScvO 2 with high precision and might serve as a faster, safer, and noninvasive surrogate to these measures.

No abstract