Artificial intelligence in digital pathology: a systematic review and meta-analysis of diagnostic test accuracy.

Ensuring diagnostic performance of artificial intelligence (AI) before introduction into clinical practice is essential. Growing numbers of studies using AI for digital pathology have been reported over recent years. The aim of this work is to examine the diagnostic accuracy of AI in digital pathology images for any disease. This systematic review and meta-analysis included diagnostic accuracy studies using any type of AI applied to whole slide images (WSIs) for any disease. The reference standard was diagnosis by histopathological assessment and/or immunohistochemistry. Searches were conducted in PubMed, EMBASE and CENTRAL in June 2022. Risk of bias and concerns of applicability were assessed using the QUADAS-2 tool. Data extraction was conducted by two investigators and meta-analysis was performed using a bivariate random effects model, with additional subgroup analyses also performed. Of 2976 identified studies, 100 were included in the review and 48 in the meta-analysis. Studies were from a range of countries, including over 152,000 whole slide images (WSIs), representing many diseases. These studies reported a mean sensitivity of 96.3% (CI 94.1-97.7) and mean specificity of 93.3% (CI 90.5-95.4). There was heterogeneity in study design and 99% of studies identified for inclusion had at least one area at high or unclear risk of bias or applicability concerns. Details on selection of cases, division of model development and validation data and raw performance data were frequently ambiguous or missing. AI is reported as having high diagnostic accuracy in the reported areas but requires more rigorous evaluation of its performance.

Reward contingency gates selective cholinergic suppression of amygdala neurons.

Basal forebrain cholinergic neurons modulate how organisms process and respond to environmental stimuli through impacts on arousal, attention, and memory. It is unknown, however, whether basal forebrain cholinergic neurons are directly involved in conditioned behavior, independent of secondary roles in the processing of external stimuli. Using fluorescent imaging, we found that cholinergic neurons are active during behavioral responding for a reward - even prior to reward delivery and in the absence of discrete stimuli. Photostimulation of basal forebrain cholinergic neurons, or their terminals in the basolateral amygdala (BLA), selectively promoted conditioned responding (licking), but not unconditioned behavior nor innate motor outputs. In vivo electrophysiological recordings during cholinergic photostimulation revealed reward-contingency-dependent suppression of BLA neural activity, but not prefrontal cortex. Finally, ex vivo experiments demonstrated that photostimulation of cholinergic terminals suppressed BLA projection neuron activity via monosynaptic muscarinic receptor signaling, while also facilitating firing in BLA GABAergic interneurons. Taken together, we show that the neural and behavioral effects of basal forebrain cholinergic activation are modulated by reward contingency in a target-specific manner.

Risk-stratified faecal immunochemical testing (FIT) for urgent colonoscopy in Lynch syndrome during the COVID-19 pandemic.

BACKGROUND: Lynch syndrome is a hereditary cancer disease resulting in an increased risk of colorectal cancer. Herein, findings are reported from an emergency clinical service implemented during the COVID-19 pandemic utilizing faecal immunochemical testing ('FIT') in Lynch syndrome patients to prioritize colonoscopy while endoscopy services were limited. METHODS: An emergency service protocol was designed to improve colonoscopic surveillance access throughout the COVID-19 pandemic in England for people with Lynch syndrome when services were extremely restricted (1 March 2020 to 31 March 2021) and promoted by the English National Health Service. Requests for faecal immunochemical testing from participating centres were sent to the National Health Service Bowel Cancer Screening South of England Hub and a faecal immunochemical testing kit, faecal immunochemical testing instructions, paper-based survey, and pre-paid return envelope were sent to patients. Reports with faecal haemoglobin results were returned electronically for clinical action. Risk stratification for colonoscopy was as follows: faecal haemoglobin less than 10 µg of haemoglobin/g of faeces (µg/g)-scheduled within 6-12 weeks; and faecal haemoglobin greater than or equal to 10 µg/g-triaged via an urgent suspected cancer clinical pathway. Primary outcomes of interest included the identification of highest-risk Lynch syndrome patients and determining the impact of faecal immunochemical testing in risk-stratified colonoscopic surveillance. RESULTS: Fifteen centres participated from June 2020 to March 2021. Uptake was 68.8 per cent amongst 558 patients invited. For 339 eligible participants analysed, 279 (82.3 per cent) had faecal haemoglobin less than 10 µg/g and 60 (17.7 per cent) had faecal haemoglobin greater than or equal to 10 µg/g. In the latter group, the diagnostic accuracy of faecal immunochemical testing was 65.9 per cent and escalation to colonoscopy was facilitated (median 49 versus 122 days, χ2 = 0.0003, P < 0.001). CONCLUSION: Faecal immunochemical testing demonstrated clinical value for Lynch syndrome patients requiring colorectal cancer surveillance during the pandemic in this descriptive report of an emergency COVID-19 response service. Further longitudinal investigation on faecal immunochemical testing efficacy in Lynch syndrome is warranted and will be examined under the 'FIT for Lynch' study (ISRCTN15740250).

Thalamus sends information about arousal but not valence to the amygdala.

RATIONALE: The basolateral amygdala (BLA) and medial geniculate nucleus of the thalamus (MGN) have both been shown to be necessary for the formation of associative learning. While the role that the BLA plays in this process has long been emphasized, the MGN has been less well-studied and surrounded by debate regarding whether the relay of sensory information is active or passive. OBJECTIVES: We seek to understand the role the MGN has within the thalamoamgydala circuit in the formation of associative learning. METHODS: Here, we use optogenetics and in vivo electrophysiological recordings to dissect the MGN-BLA circuit and explore the specific subpopulations for evidence of learning and synthesis of information that could impact downstream BLA encoding. We employ various machine learning techniques to investigate function within neural subpopulations. We introduce a novel method to investigate tonic changes across trial-by-trial structure, which offers an alternative approach to traditional trial-averaging techniques. RESULTS: We find that the MGN appears to encode arousal but not valence, unlike the BLA which encodes for both. We find that the MGN and the BLA appear to react differently to expected and unexpected outcomes; the BLA biased responses toward reward prediction error and the MGN focused on anticipated punishment. We uncover evidence of tonic changes by visualizing changes across trials during inter-trial intervals (baseline epochs) for a subset of cells. CONCLUSION: We conclude that the MGN-BLA projector population acts as both filter and transferer of information by relaying information about the salience of cues to the amygdala, but these signals are not valence-specified.

Cortical ensembles orchestrate social competition through hypothalamic outputs.

Most social species self-organize into dominance hierarchies1,2, which decreases aggression and conserves energy3,4, but it is not clear how individuals know their social rank. We have only begun to learn how the brain represents social rank5-9 and guides behaviour on the basis of this representation. The medial prefrontal cortex (mPFC) is involved in social dominance in rodents7,8 and humans10,11. Yet, precisely how the mPFC encodes relative social rank and which circuits mediate this computation is not known. We developed a social competition assay in which mice compete for rewards, as well as a computer vision tool (AlphaTracker) to track multiple, unmarked animals. A hidden Markov model combined with generalized linear models was able to decode social competition behaviour from mPFC ensemble activity. Population dynamics in the mPFC predicted social rank and competitive success. Finally, we demonstrate that mPFC cells that project to the lateral hypothalamus promote dominance behaviour during reward competition. Thus, we reveal a cortico-hypothalamic circuit by which the mPFC exerts top-down modulation of social dominance.

Urine lipoarabinomannan for rapid tuberculosis diagnosis in HIV-infected adult outpatients in Khayelitsha.

BACKGROUND: Decreasing tuberculosis (TB) mortality is constrained by diagnostic and treatment delays. The World Health Organization (WHO) recently actively recommended the point-of-care Alere Determine Lipoarabinomannan Ag assay (AlereLAM) to assist in the diagnosis of tuberculosis in specific HIV-infected outpatients. OBJECTIVES: The primary objective of this study was to compare time to ambulatory TB treatment in HIV-infected adults with CD4 ≤ 100 cells/µL before and after ('primary comparison groups') availability of AlereLAM. In pre-specified subgroups, we prospectively assessed AlereLAM-positive prevalence. METHOD: Clinicians prospectively performed AlereLAM in HIV-infected adults with TB symptoms and either CD4 ≤ 100 cells/µL or 'seriously ill' criteria. In a retrospective arm of equal duration, clinicians retrospectively collected data on HIV-infected adults with CD4 ≤ 100 cells/µL who initiated TB treatment. RESULTS: A total of 115 prospectively eligible adults (of whom 55 had CD4 ≤ 100 cells/µL) and 77 retrospectively eligible patients were included. In the primary comparison groups, the retrospective and prospective arms had similar age and sex distribution. With availability of AlereLAM, the time to TB treatment decreased from a median of 4 to 3 days (p = 0.0557). With availability of AlereLAM, same-day TB treatment initiation rose from 9.1% to 32.7% (p = 0.0006). In those with CD4 ≤ 100 only, those with 'seriously ill' criteria only, and in those meeting either, or both, of these criteria, AlereLAM was positive in 10.5%, 21.9%, 34.8% and 48.4% respectively. CONCLUSION: Availability of AlereLAM led to more patients initiating same-day TB treatment. Using both CD4 ≤ 100 and 'seriously ill' criteria gave the greatest yield. Results of this study have informed local policy design.

Acute social isolation evokes midbrain craving responses similar to hunger.

When people are forced to be isolated from each other, do they crave social interactions? To address this question, we used functional magnetic resonance imaging to measure neural responses evoked by food and social cues after participants (n = 40) experienced 10 h of mandated fasting or total social isolation. After isolation, people felt lonely and craved social interaction. Midbrain regions showed selective activation to food cues after fasting and to social cues after isolation; these responses were correlated with self-reported craving. By contrast, striatal and cortical regions differentiated between craving food and craving social interaction. Across deprivation sessions, we found that deprivation narrows and focuses the brain's motivational responses to the deprived target. Our results support the intuitive idea that acute isolation causes social craving, similar to the way fasting causes hunger.

Context-dependent plasticity of adult-born neurons regulated by cortical feedback.

In a complex and dynamic environment, the brain flexibly adjusts its circuits to preferentially process behaviorally relevant information. Here, we investigated how the olfactory bulb copes with this demand by examining the plasticity of adult-born granule cells (abGCs). We found that learning of olfactory discrimination elevates odor responses of young abGCs and increases their apical dendritic spines. This plasticity did not occur in abGCs during passive odor experience nor in resident granule cells (rGCs) during learning. Furthermore, we found that feedback projections from the piriform cortex show elevated activity during learning, and activating piriform feedback elicited stronger excitatory postsynaptic currents in abGCs than rGCs. Inactivation of piriform feedback blocked abGC plasticity during learning, and activation of piriform feedback during passive experience induced learning-like plasticity of abGCs. Our work describes a neural circuit mechanism that uses adult neurogenesis to update a sensory circuit to flexibly adapt to new behavioral demands.

Neural mechanisms of social homeostasis.

Social connections are vital to survival throughout the animal kingdom and are dynamic across the life span. There are debilitating consequences of social isolation and loneliness, and social support is increasingly a primary consideration in health care, disease prevention, and recovery. Considering social connection as an "innate need," it is hypothesized that evolutionarily conserved neural systems underlie the maintenance of social connections: alerting the individual to their absence and coordinating effector mechanisms to restore social contact. This is reminiscent of a homeostatic system designed to maintain social connection. Here, we explore the identity of neural systems regulating "social homeostasis." We review findings from rodent studies evaluating the rapid response to social deficit (in the form of acute social isolation) and propose that parallel, overlapping circuits are engaged to adapt to the vulnerabilities of isolation and restore social connection. By considering the neural systems regulating other homeostatic needs, such as energy and fluid balance, we discuss the potential attributes of social homeostatic circuitry. We reason that uncovering the identity of these circuits/mechanisms will facilitate our understanding of how loneliness perpetuates long-term disease states, which we speculate may result from sustained recruitment of social homeostatic circuits.

Dopamine enhances signal-to-noise ratio in cortical-brainstem encoding of aversive stimuli.

Dopamine modulates medial prefrontal cortex (mPFC) activity to mediate diverse behavioural functions1,2; however, the precise circuit computations remain unknown. One potentially unifying model by which dopamine may underlie a diversity of functions is by modulating the signal-to-noise ratio in subpopulations of mPFC neurons3-6, where neural activity conveying sensory information (signal) is amplified relative to spontaneous firing (noise). Here we demonstrate that dopamine increases the signal-to-noise ratio of responses to aversive stimuli in mPFC neurons projecting to the dorsal periaqueductal grey (dPAG). Using an electrochemical approach, we reveal the precise time course of pinch-evoked dopamine release in the mPFC, and show that mPFC dopamine biases behavioural responses to aversive stimuli. Activation of mPFC-dPAG neurons is sufficient to drive place avoidance and defensive behaviours. mPFC-dPAG neurons display robust shock-induced excitations, as visualized by single-cell, projection-defined microendoscopic calcium imaging. Finally, photostimulation of dopamine terminals in the mPFC reveals an increase in the signal-to-noise ratio in mPFC-dPAG responses to aversive stimuli. Together, these data highlight how dopamine in the mPFC can selectively route sensory information to specific downstream circuits, representing a potential circuit mechanism for valence processing.

Nontoxic, double-deletion-mutant rabies viral vectors for retrograde targeting of projection neurons.

Recombinant rabies viral vectors have proven useful for applications including retrograde targeting of projection neurons and monosynaptic tracing, but their cytotoxicity has limited their use to short-term experiments. Here we introduce a new class of double-deletion-mutant rabies viral vectors that left transduced cells alive and healthy indefinitely. Deletion of the viral polymerase gene abolished cytotoxicity and reduced transgene expression to trace levels but left vectors still able to retrogradely infect projection neurons and express recombinases, allowing downstream expression of other transgene products such as fluorophores and calcium indicators. The morphology of retrogradely targeted cells appeared unperturbed at 1 year postinjection. Whole-cell patch-clamp recordings showed no physiological abnormalities at 8 weeks. Longitudinal two-photon structural and functional imaging in vivo, tracking thousands of individual neurons for up to 4 months, showed that transduced neurons did not die but retained stable visual response properties even at the longest time points imaged.

A light- and calcium-gated transcription factor for imaging and manipulating activated neurons.

Activity remodels neurons, altering their molecular, structural, and electrical characteristics. To enable the selective characterization and manipulation of these neurons, we present FLARE, an engineered transcription factor that drives expression of fluorescent proteins, opsins, and other genetically encoded tools only in the subset of neurons that experienced activity during a user-defined time window. FLARE senses the coincidence of elevated cytosolic calcium and externally applied blue light, which together produce translocation of a membrane-anchored transcription factor to the nucleus to drive expression of any transgene. In cultured rat neurons, FLARE gives a light-to-dark signal ratio of 120 and a high- to low-calcium signal ratio of 10 after 10 min of stimulation. Opsin expression permitted functional manipulation of FLARE-marked neurons. In adult mice, FLARE also gave light- and motor-activity-dependent transcription in the cortex. Due to its modular design, minute-scale temporal resolution, and minimal dark-state leak, FLARE should be useful for the study of activity-dependent processes in neurons and other cells that signal with calcium.

Ionic currents influencing spontaneous firing and pacemaker frequency in dopamine neurons of the ventrolateral periaqueductal gray and dorsal raphe nucleus (vlPAG/DRN): A voltage-clamp and computational modelling study.

Dopamine (DA) neurons of the ventrolateral periaqueductal gray (vlPAG) and dorsal raphe nucleus (DRN) fire spontaneous action potentials (APs) at slow, regular patterns in vitro but a detailed account of their intrinsic membrane properties responsible for spontaneous firing is currently lacking. To resolve this, we performed a voltage-clamp electrophysiological study in brain slices to describe their major ionic currents and then constructed a computer model and used simulations to understand the mechanisms behind autorhythmicity in silico. We found that vlPAG/DRN DA neurons exhibit a number of voltage-dependent currents activating in the subthreshold range including, a hyperpolarization-activated cation current (IH), a transient, A-type, potassium current (IA), a background, 'persistent' (INaP) sodium current and a transient, low voltage activated (LVA) calcium current (ICaLVA). Brain slice pharmacology, in good agreement with computer simulations, showed that spontaneous firing occurred independently of IH, IA or calcium currents. In contrast, when blocking sodium currents, spontaneous firing ceased and a stable, non-oscillating membrane potential below AP threshold was attained. Using the DA neuron model we further show that calcium currents exhibit little activation (compared to sodium) during the interspike interval (ISI) repolarization while, any individual potassium current alone, whose blockade positively modulated AP firing frequency, is not required for spontaneous firing. Instead, blockade of a number of potassium currents simultaneously is necessary to eliminate autorhythmicity. Repolarization during ISI is mediated initially via the deactivation of the delayed rectifier potassium current, while a sodium background 'persistent' current is essentially indispensable for autorhythmicity by driving repolarization towards AP threshold.

A New Handle for a Hot Topic: Genetic Markers for Warm-Sensing.

Stepping out of an aggressively air-conditioned building into the sweltering heat evokes a number of thermoregulatory responses, both autonomic (sweating) and behavioral (peeling off a layer of clothing or seeking an iced beverage). Just as we come out of the hottest part of the summer, a study by Tan and colleagues provides an exciting breakthrough in our ability to study the neural mechanisms of keeping cool when it's hot.

Inhibitory Input from the Lateral Hypothalamus to the Ventral Tegmental Area Disinhibits Dopamine Neurons and Promotes Behavioral Activation.

Projections from the lateral hypothalamus (LH) to the ventral tegmental area (VTA), containing both GABAergic and glutamatergic components, encode conditioned responses and control compulsive reward-seeking behavior. GABAergic neurons in the LH have been shown to mediate appetitive and feeding-related behaviors. Here we show that the GABAergic component of the LH-VTA pathway supports positive reinforcement and place preference, while the glutamatergic component mediates place avoidance. In addition, our results indicate that photoactivation of these projections modulates other behaviors, such as social interaction and perseverant investigation of a novel object. We provide evidence that photostimulation of the GABAergic LH-VTA component, but not the glutamatergic component, increases dopamine (DA) release in the nucleus accumbens (NAc) via inhibition of local VTA GABAergic neurons. Our study clarifies how GABAergic LH inputs to the VTA can contribute to generalized behavioral activation across multiple contexts, consistent with a role in increasing motivational salience. VIDEO ABSTRACT.

Dorsal Raphe Dopamine Neurons Represent the Experience of Social Isolation.

The motivation to seek social contact may arise from either positive or negative emotional states, as social interaction can be rewarding and social isolation can be aversive. While ventral tegmental area (VTA) dopamine (DA) neurons may mediate social reward, a cellular substrate for the negative affective state of loneliness has remained elusive. Here, we identify a functional role for DA neurons in the dorsal raphe nucleus (DRN), in which we observe synaptic changes following acute social isolation. DRN DA neurons show increased activity upon social contact following isolation, revealed by in vivo calcium imaging. Optogenetic activation of DRN DA neurons increases social preference but causes place avoidance. Furthermore, these neurons are necessary for promoting rebound sociability following an acute period of isolation. Finally, the degree to which these neurons modulate behavior is predicted by social rank, together supporting a role for DRN dopamine neurons in mediating a loneliness-like state. PAPERCLIP.

Decoding neural circuits that control compulsive sucrose seeking.

The lateral hypothalamic (LH) projection to the ventral tegmental area (VTA) has been linked to reward processing, but the computations within the LH-VTA loop that give rise to specific aspects of behavior have been difficult to isolate. We show that LH-VTA neurons encode the learned action of seeking a reward, independent of reward availability. In contrast, LH neurons downstream of VTA encode reward-predictive cues and unexpected reward omission. We show that inhibiting the LH-VTA pathway reduces "compulsive" sucrose seeking but not food consumption in hungry mice. We reveal that the LH sends excitatory and inhibitory input onto VTA dopamine (DA) and GABA neurons, and that the GABAergic projection drives feeding-related behavior. Our study overlays information about the type, function, and connectivity of LH neurons and identifies a neural circuit that selectively controls compulsive sugar consumption, without preventing feeding necessary for survival, providing a potential target for therapeutic interventions for compulsive-overeating disorder.

Noninvasive optical inhibition with a red-shifted microbial rhodopsin.

Optogenetic inhibition of the electrical activity of neurons enables the causal assessment of their contributions to brain functions. Red light penetrates deeper into tissue than other visible wavelengths. We present a red-shifted cruxhalorhodopsin, Jaws, derived from Haloarcula (Halobacterium) salinarum (strain Shark) and engineered to result in red light-induced photocurrents three times those of earlier silencers. Jaws exhibits robust inhibition of sensory-evoked neural activity in the cortex and results in strong light responses when used in retinas of retinitis pigmentosa model mice. We also demonstrate that Jaws can noninvasively mediate transcranial optical inhibition of neurons deep in the brains of awake mice. The noninvasive optogenetic inhibition opened up by Jaws enables a variety of important neuroscience experiments and offers a powerful general-use chloride pump for basic and applied neuroscience.