Rapid and non-invasive detection of malaria parasites using near-infrared spectroscopy and machine learning

BACKGROUND

Novel and highly sensitive point-of-care malaria diagnostic and surveillance tools that are rapid and affordable are urgently needed to support malaria control and elimination.

METHODS

We demonstrated the potential of near-infrared spectroscopy (NIRS) technique to detect malaria parasites both, in vitro, using dilutions of infected red blood cells obtained from Plasmodium falciparum cultures and in vivo, in mice infected with P. berghei using blood spotted on slides and non-invasively, by simply scanning various body areas (e.g., feet, groin and ears). The spectra were analysed using machine learning to develop predictive models for infection.

FINDINGS

Using NIRS spectra of in vitro cultures and machine learning algorithms, we successfully detected low densities (<10-7 parasites/μL) of P. falciparum parasites with a sensitivity of 96% (n = 1041), a specificity of 93% (n = 130) and an accuracy of 96% (n = 1171) and differentiated ring, trophozoite and schizont stages with an accuracy of 98% (n = 820). Furthermore, when the feet of mice infected with P. berghei with parasitaemia ≥3% were scanned non-invasively, the sensitivity and specificity of NIRS were 94% (n = 66) and 86% (n = 342), respectively.

INTERPRETATION

These data highlights the potential of NIRS technique as rapid, non-invasive and affordable tool for surveillance of malaria cases. Further work to determine the potential of NIRS to detect malaria in symptomatic and asymptomatic malaria cases in the field is recommended including its capacity to guide current malaria elimination strategies.

Rapid and Non-Invasive Detection of Aedes aegypti Co-Infected with Zika and Dengue Viruses Using Near Infrared Spectroscopy

The transmission of dengue (DENV) and Zika (ZIKV) has been continuously increasing worldwide. An efficient arbovirus surveillance system is critical to designing early-warning systems to increase preparedness of future outbreaks in endemic countries. The Near Infrared Spectroscopy (NIRS) is a promising high throughput technique to detect arbovirus infection in Ae. aegypti with remarkable advantages such as cost and time effectiveness, reagent-free, and non-invasive nature over existing molecular tools for similar purposes, enabling timely decision making through rapid detection of potential disease. Our aim was to determine whether NIRS can differentiate Ae. aegypti females infected with either ZIKV or DENV single infection, and those coinfected with ZIKV/DENV from uninfected ones. Using 200 Ae. aegypti females reared and infected in laboratory conditions, the training model differentiated mosquitoes into the four treatments with 100% accuracy. DENV-, ZIKV-, and ZIKV/DENV-coinfected mosquitoes that were used to validate the model could be correctly classified into their actual infection group with a predictive accuracy of 100%, 84%, and 80%, respectively. When compared with mosquitoes from the uninfected group, the three infected groups were predicted as belonging to the infected group with 100%, 97%, and 100% accuracy for DENV-infected, ZIKV-infected, and the co-infected group, respectively. Preliminary lab-based results are encouraging and indicate that NIRS should be tested in field settings to evaluate its potential role to monitor natural infection in field-caught mosquitoes.

Malaria absorption peaks acquired through the skin of patients with infrared light can detect patients with varying parasitemia

To eliminate malaria, scalable tools that are rapid, affordable, and can detect patients with low parasitemia are required. Non-invasive diagnostic tools that are rapid, reagent-free, and affordable would also provide a justifiable platform for testing malaria in asymptomatic patients. However, non-invasive surveillance techniques for malaria remain a diagnostic gap. Here, we show near-infrared Plasmodium absorption peaks acquired non-invasively through the skin using a miniaturized hand-held near-infrared spectrometer. Using spectra from the ear, these absorption peaks and machine learning techniques enabled non-invasive detection of malaria-infected human subjects with varying parasitemia levels in less than 10 s.

First Report of the Detection of DENV 1 Virus in Human Blood Plasma with Near-Infrared Spectroscopy

Dengue virus (DENV) is the world’s most common arboviral infection, with an estimated 3.9 million people at risk of the infection, 100 million symptomatic cases and 10,000 deaths per year. Current diagnosis for DENV includes the use of molecular methods, such as polymerase chain reaction, which can be costly for routine use. The near-infrared spectroscopy (NIR) technique is a high throughput technique that involves shining a beam of infrared light on a biological sample, collecting a reflectance spectrum, and using machine learning algorithms to develop predictive algorithms. Here, we used NIR to detect DENV1 artificially introduced into whole blood, plasma, and serum collected from human donors. Machine learning algorithms were developed using artificial neural networks (ANN) and the resultant models were used to predict independent samples. DENV in plasma samples was detected with an overall accuracy, sensitivity, and specificity of 90% (N = 56), 88.5% (N = 28) and 92.3% (N = 28), respectively. However, a predictive sensitivity of 33.3% (N = 16) and 80% (N = 10) and specificity of 46.7% (N = 16) and 32% (N = 10) was achieved for detecting DENV1 in whole blood and serum samples, respectively. DENV1 peaks observed at 812 nm and 819 nm represent C-H stretch, peaks at 1130–1142 nm are related to methyl group and peaks at 2127 nm are related to saturated fatty groups. Our findings indicate the potential of NIR as a diagnostic tool for DENV, however, further work is recommended to assess its sensitivity for detecting DENV in people naturally infected with the virus and to determine its capacity to differentiate DENV serotypes and other arboviruses.

Ileal Pouch-Anal Anastomosis for Ulcerative Colitis: An Australian Institution's Experience

PURPOSE

We report outcomes and evaluate patient factors and the impact of surgical evolution on outcomes in consecutive ulcerative colitis patients who had restorative proctocolectomy with ileal pouch-anal anastomosis (IPAA) at an Australian institution over 26 years.

METHODS

Data including clinical characteristics, preoperative medical therapy, and surgical outcomes were collected. We divided eligible patients into 3 period arms (period 1, 1990 to 1999; period 2, 2000 to 2009; period 3, 2010 to 2016). Outcomes of interest were IPAA leak and pouch failure.

RESULTS

A total of 212 patients were included. Median follow-up was 50 (interquartile range, 17 to 120) months. Rates of early and late complications were 34.9% and 52.0%, respectively. Early complications included wound infection (9.4%), pelvic sepsis (8.0%), and small bowel obstruction (6.6%) while late complications included small bowel obstruction (18.9%), anal stenosis (16.8%), and pouch fistula (13.3%). Overall, IPAA leak rate was 6.1% and pouch failure rate was 4.8%. Eighty-three patients (42.3%) experienced pouchitis. Over time, we observed an increase in patient exposure to thiopurine (P=0.0025), cyclosporin (P=0.0002), and anti-tumor necrosis factor (P<0.00001) coupled with a shift to laparoscopic technique (P<0.00001), stapled IPAA (P<0.00001), J pouch configuration (P<0.00001), a modified 2-stage procedure (P=0.00012), and a decline in defunctioning ileostomy rate at time of IPAA (P=0.00002). Apart from pouchitis, there was no significant difference in surgical and chronic inflammatory pouch outcomes with time.

CONCLUSION

Despite greater patient exposure to immunomodulatory and biologic therapy before surgery coupled with a significant change in surgical techniques, surgical and chronic inflammatory pouch outcome rates have remained stable.

Rostral Anterior Cingulate Glutamine/Glutamate Disbalance in Major Depressive Disorder Depends on Symptom Severity

BACKGROUND

Patients with major depressive disorder (MDD) show glutamatergic deficits in the ventral anterior cingulate cortex. The glutamine/glutamate (Gln/Glu) ratio was proposed to be connected to glutamatergic cycling, which is hypothesized to be dysregulated in MDD. As an indicator of regional metabolite status, this ratio might be a robust state marker sensitive to clinical heterogeneity.

METHODS

Thirty-two MDD patients (mean age 40.88 ± 13.66 years, 19 women) and control subjects (mean age 33.09 ± 8.24 years, 19 women) were compared for pregenual anterior cingulate cortex levels of Gln/Glu, Gln/total creatine (tCr), Glu/tCr, and gamma-aminobutyric acid/tCr as determined by high-field magnetic resonance spectroscopy. We tested if symptom severity (Hamilton Depression Rating Scale) and anhedonia (Snaith-Hamilton Pleasure Scale) influence the relation of metabolites to clinical symptoms.

RESULTS

MDD patients showed higher Gln/Glu. This was driven by marginally higher Gln/tCr and nonsignificantly lower Glu/tCr. Groups defined by severity moderated relationship between Gln/Glu and the Hamilton Depression Rating Scale. Moreover, severe cases differed from both control subjects and moderate cases. Groups defined by the Snaith-Hamilton Pleasure Scale also displayed differential relationship between Gln/Glu and levels of anhedonia, predominantly driven by Gln/tCr.

CONCLUSIONS

We elaborate previous accounts of metabolite deficits in the anterior cingulate cortex toward increased Gln/Glu. There is a moderated relationship between severity and the ratio, which suggests consideration of different mechanisms or disease state for the respective subgroups in future studies.

Development and evaluation of a risk assessment tool to improve clinical triage accuracy for colonoscopic investigations

Background

Gastroenterology Departments at hospitals within Australia receive thousands of General Practitioner (GP)-referral letters for gastrointestinal investigations every month. Many of these requests are for colonoscopy. This study aims to evaluate the performance of the current symptoms-based triage system compared to a novel risk score using objective markers.

Methods

Patients with lower abdominal symptoms referred by their GPs and triaged by a Gastroenterology consultant to a colonoscopy consent clinic were recruited into the study. A risk assessment tool (RAT) was developed using objective data (clinical, demographic, pathology (stool test, FIT), standard blood tests and colonoscopy outcome). Colonoscopy and histology results were scored and then stratified as either significant bowel disease (SBD) or non-significant bowel disease (non-SBD).

Results

Of the 467 patients in our study, 45.1% were male, the mean age was 54.3 ± 13.8 years and mean BMI was 27.8 ± 6.2. Overall, 26% had SBD compared to 74% with non-SBD (42% of the cohort had a normal colonoscopy). Increasing severity of referral symptoms was related to a higher triage category, (rectal bleeding, P = 2.86*10^-9; diarrhoea, P = 0.026; abdominal pain, P = 5.67*10^-4). However, there was no significant difference in the prevalence of rectal bleeding (P = 0.991) or diarrhoea (P = 0.843) for SBD. Abdominal pain significantly reduced the risk of SBD (P = 0.0344, OR = 0.52, CI = 0.27-0.95). Conversely, the RAT had a very high specificity of 98% with PPV and NPV of SBD prediction, 74% and 77%, respectively. The RAT provided an odds ratio (OR) of 9.0, 95%CI 4.29-18.75, p = 2.32*10^-11), higher than the FIT test (OR = 5.3, 95%CI 2.44-11.69, p = 4.88*10^-6), blood score (OR = 2.8, 95%CI 1.72- 4.38, p = 1.47*10^-5) or age (OR = 2.5, 95%CI 1.61-4.00, 5.12*10^-5) independently. Notably, the ORs of these individual objective measures were higher than the current practice of symptoms-based triaging (OR = 1.4, 95%CI 0.88-2.11, p = 0.153).

Conclusions

It is critical that individuals with high risk of having SBD are triaged to the appropriate category with the shortest wait time. Here we provide evidence that a combination of blood markers, demographic markers and the FIT test have a higher diagnostic accuracy for SBD than FIT alone.

Richness in Functional Connectivity Depends on the Neuronal Integrity within the Posterior Cingulate Cortex

The brain's connectivity skeleton-a rich club of strongly interconnected members-was initially shown to exist in human structural networks, but recent evidence suggests a functional counterpart. This rich club typically includes key regions (or hubs) from multiple canonical networks, reducing the cost of inter-network communication. The posterior cingulate cortex (PCC), a hub node embedded within the default mode network, is known to facilitate communication between brain networks and is a key member of the "rich club." Here, we assessed how metabolic signatures of neuronal integrity and cortical thickness influence the global extent of a functional rich club as measured using the functional rich club coefficient (fRCC). Rich club estimation was performed on functional connectivity of resting state brain signals acquired at 3T in 48 healthy adult subjects. Magnetic resonance spectroscopy was measured in the same session using a point resolved spectroscopy sequence. We confirmed convergence of functional rich club with a previously established structural rich club. N-acetyl aspartate (NAA) in the PCC is significantly correlated with age (p = 0.001), while the rich club coefficient showed no effect of age (p = 0.106). In addition, we found a significant quadratic relationship between fRCC and NAA concentration in PCC (p = 0.009). Furthermore, cortical thinning in the PCC was correlated with a reduced rich club coefficient after accounting for age and NAA. In conclusion, we found that the fRCC is related to a marker of neuronal integrity in a key region of the cingulate cortex. Furthermore, cortical thinning in the same area was observed, suggesting that both cortical thinning and neuronal integrity in the hub regions influence functional integration of at a whole brain level.

Higher interference susceptibility in reaction time task is accompanied by weakened functional dissociation between salience and default mode network

BACKGROUND

The relationship between task-positive and task-negative components of brain networks has repeatedly been shown to be characterized by dissociated fluctuations of spontaneous brain activity. We tested whether the interaction between task-positive and task-negative brain areas during resting-state predicts higher interference susceptibility, i.e. increased reaction times (RTs), during an Attention Modulation by Salience Task (AMST).

METHODS

29 males underwent 3T resting-state Magnetic Resonance Imaging scanning. Subsequently, they performed the AMST, which measures RTs to early- and late-onset auditory stimuli while perceiving high- or low-salient visual distractors. We conducted seed-based resting-state functional connectivity (rsFC) analyses using global signal correction. We assessed general responsiveness and salience related interference in the AMST and set this into context of the resting-state functional connectivity (rsFC) between a key salience network region (dACC; task-positive region) and a key default mode network region (precuneus; task-negative region).

RESULTS

With increasing RTs to high- but not low-salient pictures dACC shows significantly weakened functional dissociation to a cluster in precuneus. This cluster overlaps with a cluster that correlates in its dACC rsFC with subjects' interference, as measured of high-salient RTs relative to low-salient RTs.

CONCLUSION

Our findings suggest that the interaction between salience network (SN) and default mode network (DMN) at rest predicts susceptibility to distraction. Subjects, that are more susceptible to high-salient stimuli - task-irrelevant external information - showed increased dACC rsFC toward precuneus. This is consistent with prior work in individuals with impaired attentional focus. Future studies might help to conclude whether an increased rsFC between a SN region and DMN region may serve as a predictor for clinical syndromes characterized by attentional impairments, e.g. ADHD. This could lead to an alternative, objective diagnosis and treatment of such disorders by decreasing the rsFC of these regions.

Abnormal functional global and local brain connectivity in female patients with anorexia nervosa

BACKGROUND

Previous resting-state functional connectivity studies in patients with anorexia nervosa used independent component analysis or seed-based connectivity analysis to probe specific brain networks. Instead, modelling the entire brain as a complex network allows determination of graph-theoretical metrics, which describe global and local properties of how brain networks are organized and how they interact.

METHODS

To determine differences in network properties between female patients with acute anorexia nervosa and pairwise matched healthy controls, we used resting-state fMRI and computed well-established global and local graph metrics across a range of network densities.

RESULTS

Our analyses included 35 patients and 35 controls. We found that the global functional network structure in patients with anorexia nervosa is characterized by increases in both characteristic path length (longer average routes between nodes) and assortativity (more nodes with a similar connectedness link together). Accordingly, we found locally decreased connectivity strength and increased path length in the posterior insula and thalamus.

LIMITATIONS

The present results may be limited to the methods applied during preprocessing and network construction.

CONCLUSION

We demonstrated anorexia nervosa-related changes in the network configuration for, to our knowledge, the first time using resting-state fMRI and graph-theoretical measures. Our findings revealed an altered global brain network architecture accompanied by local degradations indicating wide-scale disturbance in information flow across brain networks in patients with acute anorexia nervosa. Reduced local network efficiency in the thalamus and posterior insula may reflect a mechanism that helps explain the impaired integration of visuospatial and homeostatic signals in patients with this disorder, which is thought to be linked to abnormal representations of body size and hunger.

Dynamic disconnection of the supplementary motor area after processing of dismissive biographic narratives

INTRODUCTION

To understand the interplay between affective social information processing and its influence on mental states we investigated changes in functional connectivity (FC) patterns after audio exposure to emotional biographic narratives.

METHODS

While lying in the 7T MR scanner, 23 male participants listened to narratives of early childhood experiences of three persons, each having either a secure, dismissing, or preoccupied attachment representation. Directly after having listened to each of the prototypical narratives, participants underwent a 10-minute resting-state fMRI scan. To study changes in FC patterns between experimental conditions, three post-task conditions were compared to a baseline condition. Specific local alterations, as well as differences in connectivity patterns between distributed brain regions, were quantified using Network-based statistics (NBS) and graph metrics.

RESULTS

Using NBS, a nine-region subnetwork showing reduced FC after having listened to the dismissing narrative was identified. Of this subnetwork, only the left Supplementary Motor Area (SMA) exhibited a decrease in the nodal graph metrics degree and strength exclusively after listening to the dismissing narrative. No other region showed post-task changes in nodal metrics. A post hoc analysis of dynamic characteristics of FC of the left SMA showed a significant decrease in the dismissing condition when compared with the other conditions in the first three minutes of the scan, but faded away in the two subsequent intervals the differences.

CONCLUSIONS

Nodal metrics and NBS converge on reduced connectivity measures exclusively in left SMA in the dismissing condition, which may specifically reflect ongoing network changes underlying prolonged emotional reactivity to attachment-related processing.

Reduced functional connectivity in the thalamo-insular subnetwork in patients with acute anorexia nervosa

The neural underpinnings of anorexia nervosa (AN) are poorly understood. Results from existing functional brain imaging studies using disorder-relevant food- or body-stimuli have been heterogeneous and may be biased due to varying compliance or strategies of the participants. In this study, resting state functional connectivity imaging was used. To explore the distributed nature and complexity of brain function we characterized network patterns in patients with acute AN. Thirty-five unmedicated female acute AN patients and 35 closely matched healthy female participants underwent resting state functional magnetic resonance imaging. We used a network-based statistic (NBS) approach [Zalesky et al., 2010a] to identify differences between groups by isolating a network of interconnected nodes with a deviant connectivity pattern. Group comparison revealed a subnetwork of connections with decreased connectivity including the amygdala, thalamus, fusiform gyrus, putamen and the posterior insula as the central hub in the patient group. Results were not driven by changes in intranodal or global connectivity. No network could be identified where AN patients had increased coupling. Given the known involvement of the identified thalamo-insular subnetwork in interoception, decreased connectivity in AN patients in these nodes might reflect changes in the propagation of sensations that alert the organism to urgent homeostatic imbalances and pain-processes that are known to be severely disturbed in AN and might explain the striking discrepancy between patient's actual and perceived internal body state.