The abuse of antibiotics has caused a serious threat to human life and health. It is urgent to develop sensors that can detect multiple antibiotics quickly and efficiently. Biosensors are widely used in the field of antibiotic detection because of their high specificity. Advanced artificial intelligence/machine learning algorithms have allowed for remarkable achievements in image analysis and face recognition, but have not yet been widely used in the field of biosensors. Herein, this paper reviews the biosensors that have been widely used in the simultaneous detection of multiple antibiotics based on different detection mechanisms and biorecognition elements in recent years, and compares and analyzes their characteristics and specific applications. In particular, this review summarizes some AI/ML algorithms with excellent performance in the field of antibiotic detection, and which provide a platform for the intelligence of sensors and terminal apps portability. Furthermore, this review gives a short review of biosensors for the detection of multiple antibiotics.

In clinical diagnosis, the detection of dopamine(DA) is often interfered by epinephrine(EP), a neurotransmitter drug shares a similar structure with DA, which affects the accurate prediction of DA concentration. In this study, an electrochemiluminescence(ECL) and differential pulse voltammetry(DPV) dual model biosensor combined with machine learning algorithm was proposed for improving the selective detection of DA under EP interference. First, reduced graphene and gold nanoparticles were applied to enhance the sensitivity and stability of the sensor in ECL and DPV measurement. Compared with single model sensing, the dual model sensor could combine the superiority of ECL and DPV while improving the accuracy and reliability of experimental data. Then, nine graphic features related to concentration prediction were extracted from the ECL and DPV response curves, and pearson correlation coefficient as well as random forest algorithm were adopted for feature importance analysis. To achieve accurate signal decoupling, an intelligent biologically inspired algorithm, particle swarm optimization (PSO) algorithm, was introduced to optimize initial weights and thresholds of the traditional ANN for further model training based on multi-feature fusion, which greatly increased the prediction accuracy of DA in the presence of EP. The results demonstrated that DA could be accurately predicted in the concentration range of 9-200μM, and the R ² between the true and predicted values of DA concentration reached 0.99. The RMSE and MAE of the test dataset were 5.77μM and 4.32μM respectively. In addition, the PSO-ANN assisted ECL-DPV dual model detection scheme has been successfully applied for the determination of DA in real samples, such as goat serum and artificial urine.

Objective: This study aimed to investigate the antimicrobial susceptibilities, drug resistance mechanisms, and biofilm formation capacities of non-diphtheriae Corynebacterium strains isolated from sterile midstream urine of hospitalized patients with clinical urinary tract infections (UTIs). Methods: A total of 45 non-diphtheriae Corynebacterium isolates were recovered from sterile midstream urine. The available data of 45 patients were collected. Minimum inhibitory concentrations (MICs) of 10 commonly used antibiotics were determined. Meanwhile, the molecular resistance mechanisms of each agent were performed through PCR with specific primers. Moreover, the biofilm formation capability of each isolate on abiotic surfaces was detected with the MTT method. Results: In this study, the most prevalent three species were C. striatum (15/45, 33.3%), C. glucuronolyticum (9/45, 20.0%) and C. urealyticum (8/45, 17.8%). These three species also accounted for most renal and ureteral calculi cases. Male patients older than 50 years, especially those with underlying diseases, were more susceptible to non-diphtheriae Corynebacterium infection. All the 45 isolates were 100% susceptible to vancomycin and linezolid, but highly resistant to macrolide–lincosamide–streptogramin B (MLSB), fluoroquinolones, tetracyclines and β-lactams with corresponding mechanisms. The detection rate of multidrug–resistant (MDR) non-diphtheriae Corynebacterium is 91.1%. All isolates are able to form biofilm on abiotic surfaces, except those of C. urealyticum, C. tuberculostearicum and C. jeikeium. Isolates of C. glucuronolyticum and C. Striatum possessed the strongest biofilm formation capacity. C. amycolatum could form biofilm, but varied greatly among different isolates. Conclusion: C. striatum, C. glucuronolyticum and C. urealyticum were the most prevalent species relevant to UTIs. The high occurrence of MDR isolates and high diversities in resistance profiles, and the distinctive abilities of biofilm formation highlighted the urgency for identification to species level. We should pay more attention to the drug resistance profiles of non-diphtheriae Corynebacterium, which would help improve empirical antibiotic therapy and reduce drug resistance transmission.

In electrochemical sensing, the selectivity could be low if the analyte and interference species share similar redox potentials. This is a limiting factor in electrochemical detection of the broad-spectrum antibiotic chloramphenicol (CAP) against metronidazole (MNZ), which is commonly applied in conjunction. Herein, we present machine learning assisted detection of CAP with the presence of high concentration MNZ interference. From data collected with cyclic voltammetry (CV), differential pulse voltammetry (DPV) and chronoamperometry (CA), we quantified the features in electrochemical profiles and verified their correlation with CAP concentrations by Pearson correlation. Using correlated features, an artifcial neural network model was trained to accurately predict the concentration of CAP. The results show that the interference of MNZ could be minimized with the presented machine learning assisted electrochemical sensing scheme, and the CAP concentration could be accurately determined in buffer as well as in unmodified complex samples. We anticipate that the machine learning assisted electrochemical detection scheme could contribute to the improvement of selectivity of various electrochemical sensors.

The effect evaluation of the antibiotic susceptibility test based on bacterial solution is of great significance for clinical diagnosis and prevention of antibiotic abuse. Applying a microfluidic chip as the detection platform, the detection method of using microscopic images to observe bacteria under antibiotic can greatly speed up the detection time, which is more suitable for high-throughput detection. However, due to the influence of the depth of the microchannel, there are multiple layers of bacteria under the focal depth of the microscope, which greatly affects the counting and recognition accuracy and increases the difficulty of relocation of the target bacteria, as well as extracting the characteristics of bacterial liquid changes under the action of antibiotics. After the focal depth of the target bacteria is determined, although the z-axis can be controlled with the help of a three-dimensional micro-operator, the equipment is difficult to operate and the long-term changes of the target bacteria cannot be tracked quickly and accurately. In this paper, the YOLOv5 algorithm is adopted to accurately identify bacteria with different focusing states of multi-layer bacteria at the z-axis with any focal depth. In the meantime, a certain amount of microspheres were mixed into bacteria to assist in locating bacteria, which was convenient for tracking the growth state of bacteria over a long period, and the recognition rates of both bacteria and microspheres were high. The recognition accuracy and counting accuracy of bacteria are 0.734 and 0.714, and the two recognition rates of microspheres are 0.910 and 0.927, respectively, which are much higher than the counting accuracy of 0.142 for bacteria and 0.781 for microspheres with the method of enhanced depth of field (EDF method). Moreover, during long-term bacterial tracking and detection, target bacteria at multiple z-axis focal depth positions can be recorded by the aid of microspheres as a positioning aid for 3D reconstruction, and the focal depth positions can be repositioned within 3–10 h. The structural similarity (SSIM) of microscopic image structure differences at the same focal depth fluctuates between 0.960 and 0.975 at different times, and the root-mean-square error (RMSE) fluctuates between 8 and 12, which indicates that the method also has good relocation accuracy. Thus, this method provides the basis for rapid, high-throughput, and long-term analysis of microscopic changes (e.g., morphology, size) of bacteria detection under the addition of antibiotics with different concentrations based on microfluidic channels in the future.