Don't Fear the Artificial Intelligence: A Systematic Review of Machine Learning for Prostate Cancer Detection in Pathology.

CONTEXT.­: Automated prostate cancer detection using machine learning technology has led to speculation that pathologists will soon be replaced by algorithms. This review covers the development of machine learning algorithms and their reported effectiveness specific to prostate cancer detection and Gleason grading. OBJECTIVE.­: To examine current algorithms regarding their accuracy and classification abilities. We provide a general explanation of the technology and how it is being used in clinical practice. The challenges to the application of machine learning algorithms in clinical practice are also discussed. DATA SOURCES.­: The literature for this review was identified and collected using a systematic search. Criteria were established prior to the sorting process to effectively direct the selection of studies. A 4-point system was implemented to rank the papers according to their relevancy. For papers accepted as relevant to our metrics, all cited and citing studies were also reviewed. Studies were then categorized based on whether they implemented binary or multi-class classification methods. Data were extracted from papers that contained accuracy, area under the curve (AUC), or κ values in the context of prostate cancer detection. The results were visually summarized to present accuracy trends between classification abilities. CONCLUSIONS.­: It is more difficult to achieve high accuracy metrics for multiclassification tasks than for binary tasks. The clinical implementation of an algorithm that can assign a Gleason grade to clinical whole slide images (WSIs) remains elusive. Machine learning technology is currently not able to replace pathologists but can serve as an important safeguard against misdiagnosis.

Assessing the future global distribution of land ecosystems as determined by climate change and cropland incursion.

The geographic distribution of natural ecosystems is affected by both climate and cropland. Discussions of future land use/land cover usually focus on how cropland expands and displaces natural vegetation especially as climate change impacts become stronger. Less commonly considered is the direct influence of climate change on natural ecosystems simultaneously with cropland incursion. We combine a natural vegetation model responsive to climate with a cropland allocation algorithm to assess the relative importance of climate change compared to cropland incursion. Globally, the model indicates that climate change drives larger gains and losses than cropland incursion. For example, in the Amazonian rainforests, more than one sixth of the forest area could be lost due to climate change with cropland playing virtually no role. Our findings suggest that policies to protect specific ecosystems may be undercut by climate change and that localized analyses that fully account for the impacts of a changing climate on natural vegetation and agriculture are necessary to formulate policies that preserve natural ecosystems over the long term. Supplementary Information: The online version contains supplementary material available at 10.1007/s10584-023-03584-3.

Multi-channel attention-fusion neural network for brain age estimation: Accuracy, generality, and interpretation with 16,705 healthy MRIs across lifespan.

Brain age estimated by machine learning from T1-weighted magnetic resonance images (T1w MRIs) can reveal how brain disorders alter brain aging and can help in the early detection of such disorders. A fundamental step is to build an accurate age estimator from healthy brain MRIs. We focus on this step, and propose a framework to improve the accuracy, generality, and interpretation of age estimation in healthy brain MRIs. For accuracy, we used one of the largest sample sizes (N = 16,705). For each subject, our proposed algorithm first explicitly splits the T1w image, which has been commonly treated as a single-channel 3D image in other studies, into two 3D image channels representing contrast and morphometry information. We further proposed a "fusion-with-attention" deep learning convolutional neural network (FiA-Net) to learn how to best fuse the contrast and morphometry image channels. FiA-Net recognizes varying contributions across image channels at different brain anatomy and different feature layers. In contrast, multi-channel fusion does not exist for brain age estimation, and is mostly attention-free in other medical image analysis tasks (e.g., image synthesis, or segmentation), where treating channels equally may not be optimal. For generality, we used lifespan data 0-97 years of age for real-world utility; and we thoroughly tested FiA-Net for multi-site and multi-scanner generality by two phases of cross-validations in discovery and replication data, compared to most other studies with only one phase of cross-validation. For interpretation, we directly measured each artificial neuron's correlation with the chronological age, compared to other studies looking at the saliency of features where salient features may or may not predict age. Overall, FiA-Net achieved a mean absolute error (MAE) of 3.00 years and Pearson correlation r=0.9840 with known chronological ages in healthy brain MRIs 0-97 years of age, comparing favorably with state-of-the-art algorithms and studies for accuracy and generality across sites and datasets. We also provided interpretations on how different artificial neurons and real neuroanatomy contribute to the age estimation.

Success of Nonsedated Neuroradiologic MRI in Children 1-7 Years Old.

BACKGROUND. MRI use and the need for monitored anesthesia care (MAC) in children have increased. However, MAC is associated with examination delays, increased cost, and safety concerns. OBJECTIVE. The purpose of this study was to evaluate the success rate of nonsedated neuroradiologic MRI studies in children 1-7 years old and to investigate factors associated with success. METHODS. We retrospectively reviewed data from our institutional nonsedated MRI program. Inclusion criteria were outpatient nonsedated MRI referral, age 1-7 years old, and neuroradiologic indication. Exclusion criteria were MRI examinations for ventricular checks and contrast material use. Success was determined by reviewing the clinical MRI report. We recorded patient age and sex, type of MRI examination (brain, spine, craniospinal, head and neck, and brain with MRA), protocol length, presence of child life specialist, video goggle use, and MRI appointment time (routine daytime appointment or evening appointment). We used descriptive statistics to summarize patient demographics and clinical data and logistic regression models to evaluate predictors of success in the entire sample. Subset analyses were performed for children from 1 to < 3 years old and 3 to 7 years old. RESULTS. We analyzed 217 patients who underwent nonsedated MRI examinations (median age, 5.1 years). Overall success rate was 82.0% (n = 178). The success rates were 81.4% (n = 127) for brain, 90.3% (n = 28) for spine, 71.4% (n = 10) for craniospinal, 66.7% (n = 6) for head and neck, and 100% (n = 7) for brain with MRA. Age was significantly associated with success (odds ratio [OR], 1.33; p = .009). In children 1 to < 3 years old, none of the factors analyzed were significant predictors of success (all, p > .48). In children 3-7 years old, protocol duration (OR, 0.96; 95% CI, 0.93-0.99; p = .02) and video goggle use (OR, 6.38; 95% CI, 2.16-18.84; p = .001) were significantly associated with success. CONCLUSION. A multidisciplinary approach with age-appropriate resources enables a high success rate for nonsedated neuroradiologic MRI in children 1-7 years old. CLINICAL IMPACT. Using age as the primary criterion to determine the need for MAC may lead to overuse of these services. Dissemination of information regarding nonsedated MRI practice could reduce the rate of sedated MRI in young children.

BRAIN AGE ESTIMATION USING LSTM ON CHILDREN'S BRAIN MRI.

Brain age prediction based on children's brain MRI is an important biomarker for brain health and brain development analysis. In this paper, we consider the 3D brain MRI volume as a sequence of 2D images and propose a new framework using the recurrent neural network for brain age estimation. The proposed method is named as 2D-ResNet18+Long short-term memory (LSTM), which consists of four parts: 2D ResNet18 for feature extraction on 2D images, a pooling layer for feature reduction over the sequences, an LSTM layer, and a final regression layer. We apply the proposed method on a public multisite NIH-PD dataset and evaluate generalization on a second multisite dataset, which shows that the proposed 2D-ResNet18+LSTM method provides better results than traditional 3D based neural network for brain age estimation.

ACR Appropriateness Criteria® Head Trauma-Child.

Head trauma is a frequent indication for cranial imaging in children. The majority of accidental pediatric head trauma is minor and sustained without intracranial injury. Well-validated pediatric-specific clinical decision guidelines should be used to identify very low-risk children who can safely forgo imaging. In those who require acute imaging, CT is considered the first-line imaging modality for suspected intracranial injury because of the short duration of the examination and its high sensitivity for acute hemorrhage. MRI can accurately detect traumatic complications, but often necessitates sedation in children, owing to the examination length and motion sensitivity, which limits rapid assessment. There is a paucity of literature regarding vascular injuries in pediatric blunt head trauma and imaging is typically guided by clinical suspicion. Advanced imaging techniques have the potential to identify changes that are not seen by standard imaging, but data are currently insufficient to support routine clinical use. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision include an extensive analysis of current medical literature from peer reviewed journals and the application of well-established methodologies (RAND/UCLA Appropriateness Method and Grading of Recommendations Assessment, Development, and Evaluation or GRADE) to rate the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where evidence is lacking or equivocal, expert opinion may supplement the available evidence to recommend imaging or treatment.

ACR Appropriateness Criteria® Cerebrovascular Disease-Child.

Stroke is an uncommon but an important and under-recognized cause of morbidity and mortality in children. Strokes may be due to either brain ischemia or intracranial hemorrhage. Common symptoms of pediatric acute stroke include headache, vomiting, focal weakness, numbness, visual disturbance, seizures, and altered consciousness. Most children presenting with an acute neurologic deficit do not have an acute stroke, but have symptoms due to stroke mimics which include complicated migraine, seizures with postictal paralysis, and Bell palsy. Because of frequency of stroke mimics, in children and the common lack of specificity in symptoms, the diagnosis of a true stroke may be delayed. There are a relatively large number of potential causes of stroke mimic and true stroke. Consequently, imaging plays a critical role in the assessment of children with possible stroke and especially in children who present with acute onset of stroke symptoms. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision include an extensive analysis of current medical literature from peer reviewed journals and the application of well-established methodologies (RAND/UCLA Appropriateness Method and Grading of Recommendations Assessment, Development, and Evaluation or GRADE) to rate the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where evidence is lacking or equivocal, expert opinion may supplement the available evidence to recommend imaging or treatment.

Prenatal diagnosis of intraconal lymphatic malformation on fetal magnetic resonance imaging.

Lymphatic malformations are benign hamartomatous tumors present at birth but usually diagnosed in early childhood. We report a case of prenatal diagnosis of an isolated unilateral retrobulbar lymphatic malformation with fetal magnetic resonance imaging (MRI). This was first detected at 27 weeks' gestational age. Postnatal ocular examinations at 4 days and 5 weeks of age showed no signs of optic nerve compromise. Postnatal MRI at 18 days of age showed slight increase in size of the lesion, and no intracranial vascular malformations were detected.

Diagnostic equivalency of fast T2 and FLAIR sequences for pediatric brain MRI: a pilot study.

BACKGROUND: Faster and motion robust magnetic resonance imaging (MRI) sequences are desirable in pediatric brain MRI as they can help reduce the need for monitored anesthesia care, which is a costly and limited resource that carries medical risks. OBJECTIVE: To evaluate the diagnostic equivalency of commercially available accelerated motion robust MR sequences relative to standard sequences. MATERIALS AND METHODS: This was an institutional review board-approved prospective study. Subjects underwent a clinical brain MRI using conventional multiplanar images at 3 Tesla followed by fast axial T2 and FLAIR (fluid-attenuated inversion recovery) sequences optimized for an approximately 50% reduction in acquisition time. Conventional and fast images from each subject were reviewed by two blinded pediatric neuroradiologists. The readers evaluated the presence of 12 findings. Intra-observer agreement was estimated for fast versus conventional sequences. For each set of sequences, interobserver agreement calculations and chi-square tests were used to evaluate differences between fast and conventional acquisitions. An independent third reader reviewed the intra-observer discrepancies and adjudicated them as being more conspicuous on fast sequence, conventional sequence or the equivalent. The readers also were asked to rate motion artifacts with a previously validated score. RESULTS: Images from 77 children (mean age: 11.3 years) were analyzed. Intra-observer agreement (fast versus conventional) ranged between 89.2% and 92.3%. Interobserver agreement ranged between 86.1% and 88.4%. Interobserver agreement was significantly higher for conventional FLAIR relative to fast FLAIR for small (<5 mm) foci of T2 in the white matter. Otherwise, interobserver agreement was not different between the fast and conventional sequences. For awake subjects, fast sequences had significantly fewer artifacts (P<0.05). CONCLUSION: Conventional T2 and FLAIR sequences can be optimized to shorten acquisition while maintaining diagnostic equivalency. These faster sequences were also less susceptible to motion artifacts.

ACR Appropriateness Criteria® Scoliosis-Child.

Scoliosis is frequently encountered in childhood, with prevalence of 2%. The majority is idiopathic, without vertebral segmentation anomaly, dysraphism, neuromuscular abnormality, skeletal dysplasia, tumor, or infection. As a complement to clinical assessment, radiography is the primary imaging modality used to classify scoliosis and subsequently monitor its progression and response to treatment. MRI is utilized selectively to assess for neural axis abnormalities in those at higher risk, including those with congenital scoliosis, early onset idiopathic scoliosis, and adolescent idiopathic scoliosis with certain risk factors. CT, although not routinely employed in the initial evaluation of scoliosis, may have a select role in characterizing the bone anomalies of congenital scoliosis and in perioperative planning. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision include an extensive analysis of current medical literature from peer reviewed journals and the application of well-established methodologies (RAND/UCLA Appropriateness Method and Grading of Recommendations Assessment, Development, and Evaluation or GRADE) to rate the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where evidence is lacking or equivocal, expert opinion may supplement the available evidence to recommend imaging or treatment.

ACR Appropriateness Criteria® Suspected Spine Trauma-Child.

Choosing the appropriate imaging in children with accidental traumatic spine injuries can be challenging because the recommendations based on scientific evidence at this time differ from those applied in adults. This differentiation is due in part to differences in anatomy and physiology of the developing spine. This publication uses scientific evidence and a panel of pediatric experts to summarize best current imaging practices for children with accidental spine trauma. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision include an extensive analysis of current medical literature from peer reviewed journals and the application of well-established methodologies (RAND/UCLA Appropriateness Method and Grading of Recommendations Assessment, Development, and Evaluation or GRADE) to rate the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where evidence is lacking or equivocal, expert opinion may supplement the available evidence to recommend imaging or treatment.

Agricultural investments and hunger in Africa modeling potential contributions to SDG2 - Zero Hunger.

We use IFPRI's IMPACT framework of linked biophysical and structural economic models to examine developments in global agricultural production systems, climate change, and food security. Building on related work on how increased investment in agricultural research, resource management, and infrastructure can address the challenges of meeting future food demand, we explore the costs and implications of these investments for reducing hunger in Africa by 2030. This analysis is coupled with a new investment estimation model, based on the perpetual inventory methodology (PIM), which allows for a better assessment of the costs of achieving projected agricultural improvements. We find that climate change will continue to slow projected reductions in hunger in the coming decades-increasing the number of people at risk of hunger in 2030 by 16 million in Africa compared to a scenario without climate change. Investments to increase agricultural productivity can offset the adverse impacts of climate change and help reduce the share of people at risk of hunger in 2030 to five percent or less in Northern, Western, and Southern Africa, but the share is projected to remain at ten percent or more in Eastern and Central Africa. Investments in Africa to achieve these results are estimated to cost about 15 billion USD per year between 2015 and 2030, as part of a larger package of investments costing around 52 billion USD in developing countries.

Markerless high-frequency prospective motion correction for neuroanatomical MRI.

PURPOSE: To integrate markerless head motion tracking with prospectively corrected neuroanatomical MRI sequences and to investigate high-frequency motion correction during imaging echo trains. METHODS: A commercial 3D surface tracking system, which estimates head motion by registering point cloud reconstructions of the face, was used to adapt the imaging FOV based on head movement during MPRAGE and T2 SPACE (3D variable flip-angle turbo spin-echo) sequences. The FOV position and orientation were updated every 6 lines of k-space (< 50 ms) to enable "within-echo-train" prospective motion correction (PMC). Comparisons were made with scans using "before-echo-train" PMC, in which the FOV was updated only once per TR, before the start of each echo train (ET). Continuous-motion experiments with phantoms and in vivo were used to compare these high-frequency and low-frequency correction strategies. MPRAGE images were processed with FreeSurfer to compare estimates of brain structure volumes and cortical thickness in scans with different PMC. RESULTS: The median absolute pose differences between markerless tracking and MR image registration were 0.07/0.26/0.15 mm for x/y/z translation and 0.06º/0.02º/0.12° for rotation about x/y/z. The PMC with markerless tracking substantially reduced motion artifacts. The continuous-motion experiments showed that within-ET PMC, which minimizes FOV encoding errors during ETs that last over 1 second, reduces artifacts compared with before-ET PMC. T2 SPACE was found to be more sensitive to motion during ETs than MPRAGE. FreeSurfer morphometry estimates from within-ET PMC MPRAGE images were the most accurate. CONCLUSION: Markerless head tracking can be used for PMC, and high-frequency within-ET PMC can reduce sensitivity to motion during long imaging ETs.

ACR Appropriateness Criteria® Sinusitis-Child.

Sinusitis is common in children that usually resolves spontaneously. Imaging is not part of the standard of care for initial diagnosis, however may be necessary in cases with persistent or chronic sinusitis to guide surgical intervention, or to rule out intracranial and vascular complications of sinusitis. Computed tomography (CT) and magnetic resonance imaging (MRI) are the leading imaging modalities. In this article, appropriateness in use of imaging modalities are discussed under common/clinically relevant scenarios. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision include an extensive analysis of current medical literature from peer reviewed journals and the application of well-established methodologies (RAND/UCLA Appropriateness Method and Grading of Recommendations Assessment, Development, and Evaluation or GRADE) to rate the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where evidence is lacking or equivocal, expert opinion may supplement the available evidence to recommend imaging or treatment.

ACR Appropriateness Criteria® Headache-Child.

Headaches in children are not uncommon and have various causes. Proper neuroimaging of these children is very specific to the headache type. Care must be taken to choose and perform the most appropriate initial imaging examination in order to maximize the ability to properly determine the cause with minimum risk to the child. This evidence-based report discusses the different headache types in children and provides appropriate guidelines for imaging these children. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision include an extensive analysis of current medical literature from peer reviewed journals and the application of well-established methodologies (RAND/UCLA Appropriateness Method and Grading of Recommendations Assessment, Development, and Evaluation or GRADE) to rate the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where evidence is lacking or equivocal, expert opinion may supplement the available evidence to recommend imaging or treatment.

Meteorological Controls on Local and Regional Volcanic Ash Dispersal.

Volcanic ash has the capacity to impact human health, livestock, crops and infrastructure, including international air traffic. For recent major eruptions, information on the volcanic ash plume has been combined with relatively coarse-resolution meteorological model output to provide simulations of regional ash dispersal, with reasonable success on the scale of hundreds of kilometres. However, to predict and mitigate these impacts locally, significant improvements in modelling capability are required. Here, we present results from a dynamic meteorological-ash-dispersion model configured with sufficient resolution to represent local topographic and convectively-forced flows. We focus on an archetypal volcanic setting, Soufrière, St Vincent, and use the exceptional historical records of the 1902 and 1979 eruptions to challenge our simulations. We find that the evolution and characteristics of ash deposition on St Vincent and nearby islands can be accurately simulated when the wind shear associated with the trade wind inversion and topographically-forced flows are represented. The wind shear plays a primary role and topographic flows a secondary role on ash distribution on local to regional scales. We propose a new explanation for the downwind ash deposition maxima, commonly observed in volcanic eruptions, as resulting from the detailed forcing of mesoscale meteorology on the ash plume.

Correction to: Ionizing radiation from computed tomography versus anesthesia for magnetic resonance imaging in infants and children: patient safety considerations.

The published version of this article incorrectly lists Dr. Joseph P. Cravero in the Department of Radiology at Boston Children's Hospital. Dr. Cravero's correct affiliation is given below.

Ionizing radiation from computed tomography versus anesthesia for magnetic resonance imaging in infants and children: patient safety considerations.

In the context of health care, risk assessment is the identification, evaluation and estimation of risk related to a particular clinical situation or intervention compared to accepted medical practice standards. The goal of risk assessment is to determine an acceptable level of risk for a given clinical treatment or intervention in association with the provided clinical circumstances for a patient or group of patients. In spite of the inherent challenges related to risk assessment in pediatric cross-sectional imaging, the potential risks of ionizing radiation and sedation/anesthesia in the pediatric population are thought to be quite small. Nevertheless both issues continue to be topics of discussion concerning risk and generate significant anxiety and concern for patients, parents and practicing pediatricians. Recent advances in CT technology allow for more rapid imaging with substantially lower radiation exposures, obviating the need for anesthesia for many indications and potentially mitigating concerns related to radiation exposure. In this review, we compare and contrast the potential risks of CT without anesthesia against the potential risks of MRI with anesthesia, and discuss the implications of this analysis on exam selection, providing specific examples related to neuroblastoma surveillance imaging.

ACR Appropriateness Criteria® Back Pain-Child.

It is now generally accepted that nontraumatic back pain in the pediatric population is common. The presence of isolated back pain in a child has previously been an indication for imaging; however, recently a more conservative approach has been suggested using clinical criteria. The presence of constant pain, night pain, and radicular pain, alone or in combination, lasting for 4 weeks or more, constitute clinical red flags that should prompt further imaging. Without these clinical red flags, imaging is likely not indicated. Exceptions include an abnormal neurologic examination or clinical and laboratory findings suggesting an infectious or neoplastic etiology, and when present should prompt immediate imaging. Initial imaging should consist of spine radiographs limited to area of interest, with spine MRI without contrast to evaluate further if needed. CT of the spine, limited to area of interest, and Tc-99m bone scan whole body with single-photon emission computed tomography may be useful in some patients. The addition of intravenous contrast is also recommended for evaluation of a potential neoplastic or infectious process. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision include an extensive analysis of current medical literature from peer reviewed journals and the application of well-established methodologies (RAND/UCLA Appropriateness Method and Grading of Recommendations Assessment, Development, and Evaluation or GRADE) to rate the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where evidence is lacking or equivocal, expert opinion may supplement the available evidence to recommend imaging or treatment.