Automation of Wilms' tumor segmentation by artificial intelligence.

BACKGROUND: 3D reconstruction of Wilms' tumor provides several advantages but are not systematically performed because manual segmentation is extremely time-consuming. The objective of our study was to develop an artificial intelligence tool to automate the segmentation of tumors and kidneys in children. METHODS: A manual segmentation was carried out by two experts on 14 CT scans. Then, the segmentation of Wilms' tumor and neoplastic kidney was automatically performed using the CNN U-Net and the same CNN U-Net trained according to the OV2ASSION method. The time saving for the expert was estimated depending on the number of sections automatically segmented. RESULTS: When segmentations were performed manually by two experts, the inter-individual variability resulted in a Dice index of 0.95 for tumor and 0.87 for kidney. Fully automatic segmentation with the CNN U-Net yielded a poor Dice index of 0.69 for Wilms' tumor and 0.27 for kidney. With the OV2ASSION method, the Dice index varied depending on the number of manually segmented sections. For the segmentation of the Wilms' tumor and neoplastic kidney, it varied respectively from 0.97 to 0.94 for a gap of 1 (2 out of 3 sections performed manually) to 0.94 and 0.86 for a gap of 10 (1 section out of 6 performed manually). CONCLUSION: Fully automated segmentation remains a challenge in the field of medical image processing. Although it is possible to use already developed neural networks, such as U-Net, we found that the results obtained were not satisfactory for segmentation of neoplastic kidneys or Wilms' tumors in children. We developed an innovative CNN U-Net training method that makes it possible to segment the kidney and its tumor with the same precision as an expert while reducing their intervention time by 80%.

A review of robot-assisted ultrasound examination: Systems and technology.

BACKGROUND: At present, the number and overall level of ultrasound (US) doctors cannot meet the medical needs, and the medical ultrasound robots will largely solve the shortage of medical resources. METHODS: According to the degree of automation, the handheld, semi-automatic and automatic ultrasound examination robot systems are summarised. Ultrasound scanning path planning and robot control are the keys to ensure that the robot systems can obtain high-quality images. Therefore, the ultrasound scanning path planning and control methods are summarised. The research progress and future trends are discussed. RESULTS: A variety of ultrasound robot systems have been applied to various medical works. With the continuous improvement of automation, the systems provide high-quality ultrasound images and image guidance for clinicians. CONCLUSION: Although the development of medical ultrasound robot still faces challenges, with the continuous progress of robot technology and communication technology, medical ultrasound robot will have great development potential and broad application space.

[Application of automatic slide-dropping instrument in bone marrow chromosomal karyotyping].

OBJECTIVE: To explore the application of an automatic slide-dropping instrument in bone marrow chromosomal karyotyping. METHODS: The effects of manual and automatic dropping methods under different environmental humidity were retrospectively analyzed, and the repeatability of the automatic dropping method was analyzed. RESULTS: No statistical difference was found between the results of automatic and manual dropping methods under the optimum ambient humidity and high humidity (P > 0.05). At low humidity, there was a statistical difference between the two methods (P < 0.05). With regard to the repeatability, the coefficient of variations of the automatic dropping method for the number of split phases, the rate of good dispersion and the rate of overlap were all lower than those of the manual dropping method. A statistical difference was also found in the number of split phases (P < 0.05) but not in the discrete excellent rate and overlapping rate between the two methods (P > 0.05). CONCLUSION: Better effect can be obtained by the automatic dropping instrument. It is suggested to gradually replace manual work with machine.

Automatic localization of cochlear implant electrodes using cone beam computed tomography images.

BACKGROUND: Cochlear implants (CI) are implantable medical devices that enable the perception of sounds and the understanding of speech by electrically stimulating the auditory nerve in case of inner ear damage. The stimulation takes place via an array of electrodes surgically inserted in the cochlea. After CI implantation, cone beam computed tomography (CBCT) is used to evaluate the position of the electrodes. Moreover, CBCT is used in research studies to investigate the relationship between the position of the electrodes and the hearing outcome of CI user. In clinical routine, the estimation of the position of the CI electrodes is done manually, which is very time-consuming. RESULTS: The aim of this study was to optimize procedures of automatic electrode localization from CBCT data following CI implantation. For this, we analyzed the performance of automatic electrode localization for 150 CBCT data sets of 10 different types of electrode arrays. Our own implementation of the method by Noble and Dawant (Lecture notes in computer science (Including subseries lecture notes in artificial intelligence and lecture notes in bioinformatics), Springer, pp 152-159, 2015. https://doi.org/10.1007/978-3-319-24571-3_19 ) for automated electrode localization served as a benchmark for evaluation. Differences in the detection rate and the localization accuracy across types of electrode arrays were evaluated and errors were classified. Based on this analysis, we developed a strategy to optimize procedures of automatic electrode localization. It was shown that particularly distantly spaced electrodes in combination with a deep insertion can lead to apical-basal confusions in the localization procedure. This confusion prevents electrodes from being detected or assigned correctly, leading to a deterioration in localization accuracy. CONCLUSIONS: We propose an extended cost function for automatic electrode localization methods that prevents double detection of electrodes to avoid apical-basal confusions. This significantly increased the detection rate by 11.15 percent points and improved the overall localization accuracy by 0.53 mm (1.75 voxels). In comparison to other methods, our proposed cost function does not require any prior knowledge about the individual cochlea anatomy.

A qualitative, multi-centre approach to the current state of digitalisation and automation of surveillance in infection prevention and control in German hospitals.

BACKGROUND: Healthcare associated infections (HAI) pose a major threat to healthcare systems resulting in an increased burden of disease. Surveillance plays a key role in rapidly identifying these infections and preventing further transmissions. Alas, in German hospitals, the majority of surveillance efforts have been heavily relying on labour intensive processes like manual chart review. In order to be able to identify further starting points for future digital tools and interventions to aid the surveillance of HAI we aimed to gain an understanding of the current state of digitalisation in the context of the general surveillance organisation in German clinics across all care-levels. The end user perspective of infection prevention and control (IPC) professionals was chosen to identify digital interventions that have the biggest impact on the daily surveillance work routines of IPC professionals. Perceived impediments in the advancement of surveillance digitalisation should be explored. METHODS: Following the development of an interview guideline, eight IPC professionals from seven German hospitals of different care levels were questioned in semi- structured interviews between December 2022 and January 2023. These included questions about general surveillance organisation, access to digital data sources, software to aid the surveillance process as well as current issues in the surveillance process and implementation of software systems. Subsequently, after full transcription, the interview sections were categorized in code categories (first deductive then inductive coding) and analysed qualitatively. RESULTS: Results were characterised by high heterogeneity in terms of general surveillance organisation and access to digital data sources. Software configuration of hospital and laboratory information systems (HIS/LIS) as well as patient data management systems (PDMS) varied not only between hospitals of different care levels but also between hospitals of the same care level. Outside research projects, neither fully automatic software nor solutions utilising artificial intelligence have currently been implemented in clinical routine in any of the hospitals. CONCLUSIONS: Access to digital data sources and software is increasingly available to aid surveillance of HAI. Nevertheless, surveillance processes in hospitals analysed in this study still heavily rely on manual processes. In the analysed hospitals, there is an implementation and funding gap of (semi-) automatic surveillance solutions in clinical practice, especially in healthcare facilities of lower care levels.

Automation of systematic reviews of biomedical literature: a scoping review of studies indexed in PubMed.

BACKGROUND: The demand for high-quality systematic literature reviews (SRs) for evidence-based medical decision-making is growing. SRs are costly and require the scarce resource of highly skilled reviewers. Automation technology has been proposed to save workload and expedite the SR workflow. We aimed to provide a comprehensive overview of SR automation studies indexed in PubMed, focusing on the applicability of these technologies in real world practice. METHODS: In November 2022, we extracted, combined, and ran an integrated PubMed search for SRs on SR automation. Full-text English peer-reviewed articles were included if they reported studies on SR automation methods (SSAM), or automated SRs (ASR). Bibliographic analyses and knowledge-discovery studies were excluded. Record screening was performed by single reviewers, and the selection of full text papers was performed in duplicate. We summarized the publication details, automated review stages, automation goals, applied tools, data sources, methods, results, and Google Scholar citations of SR automation studies. RESULTS: From 5321 records screened by title and abstract, we included 123 full text articles, of which 108 were SSAM and 15 ASR. Automation was applied for search (19/123, 15.4%), record screening (89/123, 72.4%), full-text selection (6/123, 4.9%), data extraction (13/123, 10.6%), risk of bias assessment (9/123, 7.3%), evidence synthesis (2/123, 1.6%), assessment of evidence quality (2/123, 1.6%), and reporting (2/123, 1.6%). Multiple SR stages were automated by 11 (8.9%) studies. The performance of automated record screening varied largely across SR topics. In published ASR, we found examples of automated search, record screening, full-text selection, and data extraction. In some ASRs, automation fully complemented manual reviews to increase sensitivity rather than to save workload. Reporting of automation details was often incomplete in ASRs. CONCLUSIONS: Automation techniques are being developed for all SR stages, but with limited real-world adoption. Most SR automation tools target single SR stages, with modest time savings for the entire SR process and varying sensitivity and specificity across studies. Therefore, the real-world benefits of SR automation remain uncertain. Standardizing the terminology, reporting, and metrics of study reports could enhance the adoption of SR automation techniques in real-world practice.

Fully Automatic Quantitative Measurement of Equilibrium Radionuclide Angiocardiography Using a Convolutional Neural Network.

PURPOSE: The aim of this study was to generate deep learning-based regions of interest (ROIs) from equilibrium radionuclide angiography datasets for left ventricular ejection fraction (LVEF) measurement. PATIENTS AND METHODS: Manually drawn ROIs (mROIs) on end-systolic and end-diastolic images were extracted from reports in a Picture Archiving and Communications System. To reduce observer variability, preprocessed ROIs (pROIs) were delineated using a 41% threshold of the maximal pixel counts of the extracted mROIs and were labeled as ground-truth. Background ROIs were automatically created using an algorithm to identify areas with minimum counts within specified probability areas around the end-systolic ROI. A 2-dimensional U-Net convolutional neural network architecture was trained to generate deep learning-based ROIs (dlROIs) from pROIs. The model's performance was evaluated using Lin's concordance correlation coefficient (CCC). Bland-Altman plots were used to assess bias and 95% limits of agreement. RESULTS: A total of 41,462 scans (19,309 patients) were included. Strong concordance was found between LVEF measurements from dlROIs and pROIs (CCC = 85.6%; 95% confidence interval, 85.4%-85.9%), and between LVEF measurements from dlROIs and mROIs (CCC = 86.1%; 95% confidence interval, 85.8%-86.3%). In the Bland-Altman analysis, the mean differences and 95% limits of agreement of the LVEF measurements were -0.6% and -6.6% to 5.3%, respectively, for dlROIs and pROIs, and -0.4% and -6.3% to 5.4% for dlROIs and mROIs, respectively. In 37,537 scans (91%), the absolute LVEF difference between dlROIs and mROIs was <5%. CONCLUSIONS: Our 2-dimensional U-Net convolutional neural network architecture showed excellent performance in generating LV ROIs from equilibrium radionuclide angiography scans. It may enhance the convenience and reproducibility of LVEF measurements.

Automated data transfer for digital twin applications: Two case studies.

Digital twins have been gaining an immense interest in various fields over the last decade. Bringing conventional process simulation models into (near) real time are thought to provide valuable insights for operators, decision makers, and stakeholders in many industries. The objective of this paper is to describe two methods for implementing digital twins at water resource recovery facilities and highlight and discuss their differences and preferable use situations, with focus on the automated data transfer from the real process. Case 1 uses a tailor-made infrastructure for automated data transfer between the facility and the digital twin. Case 2 uses edge computing for rapid automated data transfer. The data transfer lag from process to digital twin is low compared to the simulation frequency in both systems. The presented digital twin objectives can be achieved using either of the presented methods. The method of Case 1 is better suited for automatic recalibration of model parameters, although workarounds exist for the method in Case 2. The method of Case 2 is well suited for objectives such as soft sensors due to its integration with the SCADA system and low latency. The objective of the digital twin, and the required latency of the system, should guide the choice of method. PRACTITIONER POINTS: Various methods can be used for automated data transfer between the physical system and a digital twin. Delays in the data transfer differ depending on implementation method. The digital twin objective determines the required simulation frequency. Implementation method should be chosen based on the required simulation frequency.

Automatic planning for functional lung avoidance radiotherapy based on function-guided beam angle selection and plan optimization.

Objective.This study aims to develop a fully automatic planning framework for functional lung avoidance radiotherapy (AP-FLART).Approach.The AP-FLART integrates a dosimetric score-based beam angle selection method and a meta-optimization-based plan optimization method, both of which incorporate lung function information to guide dose redirection from high functional lung (HFL) to low functional lung (LFL). It is applicable to both contour-based FLART (cFLART) and voxel-based FLART (vFLART) optimization options. A cohort of 18 lung cancer patient cases underwent planning-CT and SPECT perfusion scans were collected. AP-FLART was applied to generate conventional RT (ConvRT), cFLART, and vFLART plans for all cases. We compared automatic against manual ConvRT plans as well as automatic ConvRT against FLART plans, to evaluate the effectiveness of AP-FLART. Ablation studies were performed to evaluate the contribution of function-guided beam angle selection and plan optimization to dose redirection.Main results.Automatic ConvRT plans generated by AP-FLART exhibited similar quality compared to manual counterparts. Furthermore, compared to automatic ConvRT plans, HFL mean dose,V20, andV5were significantly reduced by 1.13 Gy (p< .001), 2.01% (p< .001), and 6.66% (p< .001) respectively for cFLART plans. Besides, vFLART plans showed a decrease in lung functionally weighted mean dose by 0.64 Gy (p< .01),fV20by 0.90% (p= 0.099), andfV5by 5.07% (p< .01) respectively. Though inferior conformity was observed, all dose constraints were well satisfied. The ablation study results indicated that both function-guided beam angle selection and plan optimization significantly contributed to dose redirection.Significance.AP-FLART can effectively redirect doses from HFL to LFL without severely degrading conventional dose metrics, producing high-quality FLART plans. It has the potential to advance the research and clinical application of FLART by providing labor-free, consistent, and high-quality plans.

A Novel Apparatus for the Fully Automated Extraction and Online Liquid Chromatographic Analysis of Solid Environmental Samples: Application to the Pressurized Hot Water Extraction of Pharmaceuticals in Soil.

A new self-assembled apparatus for the extraction of solid samples was designed and implemented to perform a recirculated pressurized hot water extraction (R-PHWE) directly coupled to liquid chromatography-tandem mass spectrometry. To investigate the potential of this new extraction apparatus, 34 target pharmaceutical compounds were analyzed in loam, silt-loam, and silty-clay-loam soils. The target analytes were characterized by heterogeneous physicochemical properties (e.g., -1.60 ≤ log D ≤ 5.91 at pH = 7.2, i.e., at the mean pH values of the three soils). Design of experiments (DoE) was used to identify the best extraction conditions for the target analytes by studying temperature, pressure, and number of extraction cycles. The results of DoE optimization pointed out the significant influence of the number of cycles on recovery. The application of DoE set point to the three reference soils provided recoveries ≥60% for 21-25 out the 34 target analytes, depending on soil. Good recovery precision (<25%) and moderate suppressive matrix effect (≤40%) were found for most target analytes, regardless of the soil considered. The optimized R-PHWE procedure evidenced statistically higher recoveries for 16 out of 34 target analytes when compared to conventional off-line dynamic PHWE.

Hetero-Diels-Alder and CuAAC Click Reactions for Fluorine-18 Labeling of Peptides: Automation and Comparative Study of the Two Methods.

Radiolabeled peptides are valuable tools for diagnosis or therapies; they are often radiofluorinated using an indirect approach based on an F-18 prosthetic group. Herein, we are reporting our results on the F-18 radiolabeling of three peptides using two different methods based on click reactions. The first one used the well-known CuAAC reaction, and the second one is based on our recently reported hetero-Diels-Alder (HDA) using a dithioesters (thia-Diels-Alder) reaction. Both methods have been automated, and the 18F-peptides were obtained in similar yields and synthesis time (37-39% decay corrected yields by both methods in 120-140 min). However, to obtain similar yields, the CuAAC needs a large amount of copper along with many additives, while the HDA is a catalyst and metal-free reaction necessitating only an appropriate ratio of water/ethanol. The HDA can therefore be considered as a minimalist method offering easy access to fluorine-18 labeled peptides and making it a valuable additional tool for the indirect and site-specific labeling of peptides or biomolecules.

Drinking and driving: A systematic review of the impacts of alcohol consumption on manual and automated driving performance.

INTRODUCTION: Almost a third of car accidents involve driving after alcohol consumption. Autonomous vehicles (AVs) may offer accident-prevention benefits, but at current automation levels, drivers must still perform manual driving tasks when automated systems fail. Therefore, understanding how alcohol affects driving in both manual and automated contexts offers insight into the role of future vehicle design in mediating crash risks for alcohol-impaired driving. METHOD: This study conducted a systematic review on alcohol effects on manual and automated (takeover) driving performance. Fifty-three articles from eight databases were analyzed, with findings structured based on the information processing model, which can be extended to the AV takeover model. RESULTS: The literature indicates that different Blood Alcohol Concentration (BAC) levels affect driving skills essential for traffic safety at various information processing stages, such as delayed reacting time, impaired cognitive abilities, and hindered execution of driving tasks. Additionally, the driver's driving experience, drinking habits, and external driving environment play important roles in influencing driving performance. CONCLUSIONS: Future work is needed to examine the effects of alcohol on driving performance, particularly in AVs and takeover situations, and to develop driver monitoring systems. PRACTICAL APPLICATIONS: Findings from this review can inform future experiments, AV technology design, and the development of driver state monitoring systems.

Audiovisual presentation of variable message signs improves message processing in distracted drivers during partially automated driving.

INTRODUCTION: Highly automated driving is expected to reduce the accident risk occurrence by human errors, but it can also increase driver distraction. Previous evidence shows that auditory signals can help drivers take over in critical situations. However, it is still uncertain whether the potential benefit of verbal auditory signals could be generalized to driving situations where drivers are visually and auditorily distracted. METHOD: Our first objective was to compare the effectiveness of complementary audio messages (audio + visual condition) and visual only (visual condition) variable message signs (VMS) messages. The second objective was to explore the potential use of oral messages with traffic information to help highly-automated vehicle drivers identify critical situations. Eye-tracking data were also registered. Twenty-four volunteers participated in a driving simulator study, completing two tasks: (a) a TV series task, where they had to pay attention to an episode of a TV series while traveling along the route; and (b) a VMS task, where they had to recover the manual control of the car if the VMS message was a 'critical message.' RESULTS: General results showed that, when the audio was available, the participants: (a) had a higher ability to discriminate the VMS messages, (b) were less conservative, (c) responded earlier, and (d) their pattern of fixations was more efficient. A complementary analysis showed that the counterbalance order was a moderating factor for the discrimination ability and the response distance measures. This evidence suggests a potential learning effect, not cancelled by counterbalancing the order of the conditions. CONCLUSION: The processing of traffic messages may improve when provided as oral and visual messages. PRACTICAL APPLICATIONS: These results would be of special interest for engineers designing highly automated cars, considering that the design of automated systems must ensure that the driver's attention is sufficient to take over control.

A Microfluidic Chip-Based Automated System for Whole-Course Monitoring the Drug Responses of Organoids.

Tumor patients-derived organoids, as a promising preclinical prediction model, have been utilized to evaluate ex vivo drug responses for formulating optimal therapeutic strategies. Detecting adenosine triphosphate (ATP) has been widely used in existing organoid-based drug response tests. However, all commercial ATP detection kits containing the cell lysis procedure can only be applied for single time point ATP detection, resulting in the neglect of dynamic ATP variations in living cells. Meanwhile, due to the limited number of viable organoids from a single patient, it is impractical to exhaustively test all potential time points in search of optimal ones. In this work, a multifunctional microfluidic chip was developed to perform all procedures of organoid-based drug response tests, including establishment, culturing, drug treatment, and ATP monitoring of organoids. An ATP sensor was developed to facilitate the first successful attempt on whole-course monitoring the growth status of fragile organoids. To realize a clinically applicable automatic system for the drug testing of lung cancer, a microfluidic chip based automated system was developed to perform entire organoid-based drug response test, bridging the gap between laboratorial manipulation and clinical practices, as it outperformed previous methods by improving data repeatability, eliminating human error/sample loss, and more importantly, providing a more accurate and comprehensive evaluation of drug effects.

Automated high throughput IgG N-glycosylation sample preparation method development on the Tecan Freedom EVO platform.

Introduction: Glycomics, focusing on the role of glycans in biological processes, particularly their influence on the folding, stability and receptor interactions of glycoconjugates like antibodies, is vital for our understanding of biology. Changes in immunoglobulin G (IgG) N-glycosylation have been associated with various physiological and pathophysiological conditions. Nevertheless, time-consuming manual sample preparation is one of the limitations in the glycomics diagnostic implementation. The study aimed to develop an automated method for sample preparation on the Tecan Freedom Evo 200 platform and compare its efficiency and precision with the manual counterpart. Materials and methods: The initial method development included 32 pooled blood plasma technical replicates. An additional 24 pooled samples were used in the method comparison along with 78 random duplicates of plasma samples collected from 10,001 Dalmatians biobank to compare the manual and automated methods. Results: The development resulted in a new automated method. For the automated method, glycan peaks comprising 91% of the total sample glycan showed a variation of less than 5% while 92% of the total sample showed a variation of less than 5% for the manual method. The results of the Passing-Bablok regression indicated no differences between the automated and manual methods for 12 glycan peaks (GPs). However, for 8 GPs systematic difference was present, while both systematic and proportional differences were present for four GPs. Conclusions: The developed automated sample preparation method for IgG glycan analysis reduced exposure to hazardous chemicals and offered a simplified workflow. Despite slight differences between the methods, the new automated method showed high precision and proved to be highly comparable to its manual counterpart.

Automated diagnosis of schizophrenia based on spatial-temporal residual graph convolutional network.

BACKGROUND: Schizophrenia (SZ), a psychiatric disorder for which there is no precise diagnosis, has had a serious impact on the quality of human life and social activities for many years. Therefore, an advanced approach for accurate treatment is required. NEW METHOD: In this study, we provide a classification approach for SZ patients based on a spatial-temporal residual graph convolutional neural network (STRGCN). The model primarily collects spatial frequency features and temporal frequency features by spatial graph convolution and single-channel temporal convolution, respectively, and blends them both for the classification learning, in contrast to traditional approaches that only evaluate temporal frequency information in EEG and disregard spatial frequency features across brain regions. RESULTS: We conducted extensive experiments on the publicly available dataset Zenodo and our own collected dataset. The classification accuracy of the two datasets on our proposed method reached 96.32% and 85.44%, respectively. In the experiment, the dataset using delta has the best classification performance in the sub-bands. COMPARISON WITH EXISTING METHODS: Other methods mainly rely on deep learning models dominated by convolutional neural networks and long and short time memory networks, lacking exploration of the functional connections between channels. In contrast, the present method can treat the EEG signal as a graph and integrate and analyze the temporal frequency and spatial frequency features in the EEG signal. CONCLUSION: We provide an approach to not only performs better than other classic machine learning and deep learning algorithms on the dataset we used in diagnosing schizophrenia, but also understand the effects of schizophrenia on brain network features.

Towards more precise automatic analysis: a systematic review of deep learning-based multi-organ segmentation.

Accurate segmentation of multiple organs in the head, neck, chest, and abdomen from medical images is an essential step in computer-aided diagnosis, surgical navigation, and radiation therapy. In the past few years, with a data-driven feature extraction approach and end-to-end training, automatic deep learning-based multi-organ segmentation methods have far outperformed traditional methods and become a new research topic. This review systematically summarizes the latest research in this field. We searched Google Scholar for papers published from January 1, 2016 to December 31, 2023, using keywords "multi-organ segmentation" and "deep learning", resulting in 327 papers. We followed the PRISMA guidelines for paper selection, and 195 studies were deemed to be within the scope of this review. We summarized the two main aspects involved in multi-organ segmentation: datasets and methods. Regarding datasets, we provided an overview of existing public datasets and conducted an in-depth analysis. Concerning methods, we categorized existing approaches into three major classes: fully supervised, weakly supervised and semi-supervised, based on whether they require complete label information. We summarized the achievements of these methods in terms of segmentation accuracy. In the discussion and conclusion section, we outlined and summarized the current trends in multi-organ segmentation.

Development of automated metabolite control using mid-infrared probe for bioprocesses and vaccine manufacturing.

Automation of metabolite control in fermenters is fundamental to develop vaccine manufacturing processes more quickly and robustly. We created an end-to-end process analytical technology and quality by design-focused process by replacing manual control of metabolites during the development of fed-batch bioprocesses with a system that is highly adaptable and automation-enabled. Mid-infrared spectroscopy with an attenuated total reflectance probe in-line, and simple linear regression using the Beer-Lambert Law, were developed to quantitate key metabolites (glucose and glutamate) from spectral data that measured complex media during fermentation. This data was digitally connected to a process information management system, to enable continuous control of feed pumps with proportional-integral-derivative controllers that maintained nutrient levels throughout fed-batch stirred-tank fermenter processes. Continuous metabolite data from mid-infrared spectra of cultures in stirred-tank reactors enabled feedback loops and control of the feed pumps in pharmaceutical development laboratories. This improved process control of nutrient levels by 20-fold and the drug substance yield by an order of magnitude. Furthermore, the method is adaptable to other systems and enables soft sensing, such as the consumption rate of metabolites. The ability to develop quantitative metabolite templates quickly and simply for changing bioprocesses was instrumental for project acceleration and heightened process control and automation. ONE-SENTENCE SUMMARY: Intelligent digital control systems using continuous in-line metabolite data enabled end-to-end automation of fed-batch processes in stirred-tank reactors.